JPWO2018088403A1 - Fragment antibody and protein crystallization method using the fragment antibody - Google Patents

Abstract

The present invention can be easily produced as having antigen-binding activity, and even if it can be obtained in an E. coli expression system, it is more complex with Fv-class (v1) and antigen molecules. The present invention, which aims to provide a fragment antibody with high crystallization ability, is a peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the heavy chain domain (VH region) of the antibody, and the VH region is determined by the Chothia method. A peptide in which the amino acid residue of antibody residue number 112 was mutated to cysteine (VH (112C) -SARAH), and a peptide in which the N-terminus of the SARAH domain was bound to the C-terminus of the light chain domain (VL region) of the antibody. Thus, a peptide (VL-SARA (3 C)) relating fragment antibody consisting complex with.

Description

  The present invention relates to a fragment antibody and a protein crystallization method using the fragment antibody.

  Antibody molecules such as IgG are now indispensable tools in the fields of research, therapy, and diagnosis because of their ability to bind antigen molecules. An antibody molecule is generally a multidomain glycoprotein with a large molecular weight, and since it has a plurality of disulfide bonds (hereinafter sometimes abbreviated as SS bonds), it must be expressed using animal cells. . However, protein production using animal cells generally costs more than production using bacteria, and recombinant antibody molecules are often poorly expressed, resulting in low productivity.

  In order to compensate for the above problems in the production of antibody molecules, the use of fragment antibodies having a small molecular weight has been devised. Fragment antibodies have advantages such as high tissue invasiveness.

  Representative fragment antibodies include Fabs composed of variable domains (VH region and VL region) and constant domains (CH1 region and CL region). However, Fab is generally obtained by purifying a full-length antibody, then treating it with an enzyme and then further purifying it, so that there is a problem that the production cost is high. As another fragment antibody, a single-chain Fv (hereinafter sometimes abbreviated as scFv) in which the C-terminal of the VH region or VL region and the other N-terminus are combined with a long and flexible peptide linker is known. (Non-Patent Document 1). However, due to the presence of a long and flexible peptide linker, aggregation of scFvs may occur. In addition, there is a problem that the activity as an antibody is lowered or destabilized.

  In order to solve the problems of conventional fragment antibodies, the inventors of the human Mammalian sterile 20-like kinase 1 (Mst1) form an antiparallel coiled coil between the C-terminals of the VH region and the VL region. A new fragment antibody (hereinafter sometimes abbreviated as Fv-class first generation or Fv-class (v1)) linked through a dimer of the SARAH domain was developed (Non-patent Document 2).

  This Fv-clasp (v1) is a peptide in which the N terminus of the SARAH domain of human Mst1 is bound to the C terminus of the heavy chain domain (VH region) of the antibody, and the N terminus of the SARAH domain of the human Mst1 A peptide in which the 35th amino acid residue is mutated to cysteine, and (b) a peptide in which the N-terminus of the SARAH domain of human Mst1 is bound to the C-terminus of the light chain domain (VL region) of the antibody, A fragment antibody comprising a complex with a peptide in which the 24th amino acid residue from the N-terminal of the SARAH domain of Mst1 is mutated to cysteine, wherein (c) two SARAH domains have disulfide bonds between mutated cysteines Fragment antibody bound by

Robert E. Bird, 1988, Science, James S. Huston, 1988, PNAS Takao Arimori, Atsushi Machida, Yuki Fujii, Atsushi Kitago, Junichi Takagi, Design of New Fragment Antibody Format “Fv-class” and Its Structural Analysis, 15th Annual Meeting of the Protein Science Society of Japan, June 25, 2015 Chothia and Lesk, Canonical structures for the hypervariable regions of immunoglobulins, J. et al. Mol. Biol. , 196, 901-917 (1987) Al-Lazikani, B.I. Lesk, A .; M.M. & Chothia, C.I. (1997). Standard informations for the canonical structures of immunoglobulins. J. et al. Mol. Biol. 273, 927-948

  The inventors of the present invention have expressed and purified Fv-class (v1) against various target antigen molecules by gene recombination using animal cells or E. coli, crystallization of Fv-class (v1) and Fv-class (v1). ) And the target antigen molecule were crystallized. As a result, especially in the case of Fv-class (v1) obtained in the E. coli expression system, crystals of a complex of Fv-class (v1) and the target antigen molecule cannot be obtained, or crystals with low resolution can be obtained. It turns out that there is.

  Therefore, the present invention can be easily produced as having antigen-binding activity, and even if it is obtained in an E. coli expression system, it is a crystal of a complex of itself and an antigen molecule rather than Fv-class (v1). It is an object of the present invention to provide a fragment antibody having a high chemical ability.

In order to solve the above problems, the present inventors have (a) a peptide in which the N-terminus of the SARAH domain of human Mst1 is bound to the C-terminus of the heavy chain domain (VH region) of the antibody (hereinafter abbreviated as VH-SARAH). A peptide in which a cysteine mutation is introduced into the VH region in (a), and (b) a peptide in which the N-terminus of the SARAH domain of human Mst1 is bound to the C-terminus of the light chain domain (VL region) of the antibody (hereinafter, VL-SARAH (in some cases, abbreviated as VL-SARAH) Research on crystallization of a complex of a fragment antibody and an antigen molecule with a plurality of fragment antibodies consisting of a complex with a peptide having a cysteine mutation in the SARAH domain did.
As a result, (a) a peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the heavy chain domain (VH region) of the antibody, the VH region is converted to the amino acid residue of antibody residue number 112 by the Chothia method as cysteine. A mutated peptide (hereinafter sometimes abbreviated as VH (112C) -SARAH), and (b) a peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody. A fragment antibody comprising a complex with a peptide in which the 13th amino acid residue from the C-terminal of the SARAH domain is mutated to cysteine (hereinafter sometimes abbreviated as VL-SARAH (37C)), (c ) VH (112C) -SARAH and VL-SARAH (37C) are linked by a disulfide bond between the two cysteines. Fv-class (v2) obtained in an E. coli expression system using a fragment antibody (hereinafter sometimes abbreviated as Fv-class second generation or Fv-class (v2)). It has been found that it has the ability to crystallize itself or a complex of Fv-clasp (v2) and an antigen molecule.
In addition, it has also been found that Fv-class (v2) can obtain high-resolution crystals of a complex of Fv-class (v2) and an antigen molecule from the screening stage prior to optimization of crystallization conditions. . Further, since crystallization of hardly crystallized proteins was promoted, it was also found that Fv-clasp (v2) works as a fragment antibody for promoting protein crystallization. Furthermore, it has been found that Fv-class (v2) has higher thermal stability than Fv-class (v1) and scFv, and the present invention has been completed.

The present invention relates to the following fragment antibodies and protein crystallization methods using the fragment antibodies.
[1] (a) A peptide in which the N-terminus of the SARAH domain is linked to the C-terminus of the heavy chain domain (VH region) of the antibody, wherein the VH region has the amino acid residue of antibody residue number 112 by the Chothia method as cysteine A mutated peptide (VH (112C) -SARAH), and (b) a peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody. A fragment antibody consisting of a complex with a peptide in which the amino acid residue is mutated to cysteine (VL-SARAH (37C)), (c) VH (112C) -SARAH and VL-SARAH (37C), A fragment antibody bound by a disulfide bond between the two cysteines.
[2] The SARAH domain in the VH (112C) -SARAH is selected from those shown by SEQ ID NOs: 1-8 (shown by SEQ ID NOs: 1-8), and in the VL-SARAH (37C) The fragment antibody according to [1], wherein the SARAH domain is represented by SEQ ID NOs: 9 to 16.
[3] The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1-2, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9-10. [1] The fragment antibody according to [1].
[4] (a) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the heavy chain domain (VH region) of the antibody, wherein the VH region is the amino acid residue of antibody residue number 112 by the Chothia method as cysteine A mutated peptide (VH (112C) -SARAH), and (b) a peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody. A fragment antibody for promoting protein crystallization, comprising a complex with a peptide in which the amino acid residue is mutated to cysteine (VL-SARAH (37C)), (c) VH (112C) -SARAH and VL-SARAH (37C) is a protein crystallization-promoting protein which is bound by a disulfide bond between the two cysteines. Segment antibody.
[5] The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1 to 8, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9 to 16. [4] A fragment antibody for promoting protein crystallization according to [4].
[6] The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1-2, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9-10. [4] A fragment antibody for promoting protein crystallization according to [4].
[7] A protein crystallization method using the fragment antibody represented by the above [1] to [3].

[1] The fragment antibody of the present invention can be easily produced and has antigen binding activity.
[2] Even if the fragment antibody of the present invention is obtained in an E. coli expression system, it has a high ability to crystallize itself and a complex with an antigen molecule. Thereby, the fragment antibody of the present invention can easily analyze the three-dimensional structure of the antigen-determining site, which has been difficult with conventional fragment antibodies and antibodies.
[3] Furthermore, the fragment antibody of the present invention promotes crystallization of a protein, particularly a complex with a hardly crystallized protein, and acts as a fragment antibody for promoting protein crystallization. In addition, according to the fragment antibody of the present invention, a crystal with a complex with a protein can be obtained in a short period with good reproducibility.
[4] The fragment antibody of the present invention is a more stable fragment antibody having higher heat resistance than conventional antibodies and fragment antibodies.

FIG. 1 is a diagram confirming that P20.1 antibody-Fv-clasp (v1) was obtained by SDS-PAGE in Comparative Example 1. FIG. 2 is a diagram confirming that P20.1 antibody-Fv-clasp (v1) was obtained by SDS-PAGE in Comparative Example 2. FIG. 3 is a diagram showing that 12CA5 antibody-Fv-clasp (v1) was obtained by SDS-PAGE in Comparative Example 3. FIG. 4 shows the results of X-ray crystal structure analysis of a complex of Pv0.1 antibody Fv-clasp (v1) and C8 peptide performed in Comparative Example 5. FIG. 5 shows the result of X-ray crystal structure analysis of a complex of 12v5 antibody Fv-clasp (v1) and HA peptide performed in Comparative Example 6. FIG. 6 shows Fv-class (no SS) of 2H5 antibody, Fv-class (v1 ′) of 2H5 antibody, and Fv of 2H5 antibody obtained in Comparative Example 7-9 and Example 1 in Experimental Example 1. It is the figure which confirmed the SS bond formation ability of Fv-clasp (v2) of -clasp (v1 '') and 2H5 antibody by SDS-PAGE. FIG. 7 is a diagram showing that P20.1 antibody-Fv-clasp (v2) was obtained by SDS-PAGE in Example 2. FIG. 8 is a diagram showing that various Fv-class (v2) were obtained by SDS-PAGE in Example 3-10. FIG. 9 is a diagram showing that various Fv-clasps (v2) were confirmed to have antigen-binding activity by gel filtration chromatography in Experimental Example 2. FIG. 10 is a diagram in which it was confirmed in Experiment 2 that various Fv-clasps (v2) have antigen binding activity by biolayer interferometry. FIG. 11 (A) is a photomicrograph of the crystals obtained for TS2 / 16 antibody-Fv-clasp (v2) in Example 11. FIG. 11B shows the result of X-ray crystal structure analysis of TS2 / 16 antibody-Fv-clasp (v2) performed in Example 11. FIG. 12 (A) is a photomicrograph of crystals obtained in Example 12 for the complex of P20.1 antibody-Fv-clasp (v2) and antigen molecule. FIG. 12B shows the results of X-ray crystal structure analysis of a complex of P20.1 antibody-Fv-class (v2) and an antigen molecule performed in Example 12. FIG. 13 (A) is a micrograph of crystals obtained in Example 13 for a complex of 12CA5 antibody-Fv-clasp (v2) and an antigen molecule. FIG. 13B shows the results of X-ray crystal structure analysis of a complex of 12CA5 antibody-Fv-class (v2) and an antigen molecule performed in Example 13. FIG. 14 (A) is a photomicrograph of the crystals obtained in Example 14 for the complex of NZ-1 antibody-Fv-clasp (v2) and antigen molecule. FIG. 14B shows the result of X-ray crystal structure analysis of a complex of NZ-1 antibody-Fv-class (v2) and an antigen molecule performed in Example 14. FIGS. 15A and 15B are photomicrographs of the crystals obtained for the complex of 93201 antibody-Fv-clasp (v2) and Vps10p in Example 15. FIG. FIG. 15C shows the result of X-ray crystal structure analysis of a complex of 93201 antibody-Fv-class (v2) and Vps10p performed in Example 15. FIGS. 16A and 16B are photomicrographs of crystals obtained in Example 16 for the complex of TS2 / 16 antibody-Fv-clasp (v2) and integrin α6β1. FIG. 16C shows the result of X-ray crystal structure analysis of the complex of TS2 / 16 antibody-Fv-clasp (v2) and integrin α6β1 performed in Example 16. FIG. 17A is a photomicrograph of the crystals obtained in Example 17 for the complex of t1E4 antibody-Fv-clasp (v2) and HGF. FIG. 17B shows the result of X-ray crystal structure analysis of a complex of t1E4 antibody-Fv-class (v2) and HGF performed in Example 17.

The fragment antibody of the present invention comprises
(A) A peptide in which the N-terminus of the SARAH domain is linked to the C-terminus of the antibody heavy chain domain (VH region), and the amino acid residue of antibody residue number 112 by the Chothia method is mutated to cysteine in the VH region A peptide (VH (112C) -SARAH);
(B) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody, wherein the 13th amino acid residue from the C-terminus of the SARAH domain is mutated to cysteine (VL -A fragment antibody consisting of a complex with SARAH (37C)),
(C) A fragment antibody in which VH (112C) -SARAH and VL-SARAH (37C) are bound by a disulfide bond between the two cysteines.
Non-patent documents 3 and 4 are documents relating to a method for determining antibody residue numbers by the Chothia method.

[VH (112C) -SARAH]
The SARAH domain is usually 42 to 54, preferably 43 to 49, more preferably 47 to 49, particularly preferably composed of a short helix (h1) on the N-terminal side and a long helix (h2) on the C-terminal side. Is a domain (peptide) consisting of 49 amino acid residues, and has the property of forming an antiparallel coiled coil between each h2 and another SARAH domain. H1 preferably consists of 5 to 7 amino acids, and h2 preferably consists of 38 to 42 amino acid residues.
The SARAH domains used here may be those having the above-described properties. Among them, when two SARAH domains form an antiparallel coiled coil between h2, the two SARAH domains (two h1 ) The distance between both N terminals is preferably 35 to 45 mm, more preferably 39 to 41 mm, and particularly preferably 40 mm.
Specific examples of such SARAH domains include, for example, the SARAH domain of human Mammalian sterile 20-like kinase 1 (Mst1), the SARAH domain of human Mammalian sterile 20-like kinase 2 (Mst2), and the SRAH domain of Drosophila. Domain, the SARAH domain of Drosophila RASSF, the SARAH domain of human RASSF5, the SARAH domain of human RASSF1, the SARAH domain of human WW45, the SARAH domain of Drosophila Sav, and one of these (preferably SARAH of human Mst1). Domain) and usually more than 85% sequence homology, preferably more than 90% sequence homology, more preferred It can be mentioned, such as those having a sequence homology of 95% or more.
More specifically, for example, one represented by SEQ ID NO: 1 (SARAH domain of Mst1), one represented by SEQ ID NO: 2 (SARAH domain of Mst2), one represented by SEQ ID NO: 3 (SARAH domain of Hippo), Those shown by SEQ ID NO: 4 (RASSF domain of RASSF), those shown by SEQ ID NO: 5 (SARAH domain of RASSF5), those shown by SEQ ID NO: 6 (SARAH domain of RASSF1), those shown by SEQ ID NO: 7 ( WW45 SARAH domain), those represented by SEQ ID NO: 8 (Sav SARAH domain) and the like are preferred, those represented by SEQ ID NOS: 1-2 are more preferred, and those represented by SEQ ID NO: 1 are particularly preferred. In addition, as long as it has an above-described property, what is shown by sequence number 1-8 may be a thing by which the 1 or several amino acid in the arrangement | sequence was deleted, substituted, or added. Here, several represents a natural number of 5 or less, preferably 3 or less, more preferably 2 or less. As such, for example, in SEQ ID NOs: 1 to 8, the 26th amino acid residue from the C-terminus is substituted (mutated) with cysteine.
The SARAH domain may be either a SARAH domain derived from a naturally occurring protein or an artificially designed SARAH domain derived from a naturally occurring protein.
A SARAH domain derived from a naturally occurring protein is usually used regardless of the species having the protein. However, when the fragment antibody of the present invention is used as a medicine, it is administered in actual use from the viewpoint of antigenicity. Those derived from the same species as the animal to be used are preferred.

The VH region used here has at least a site sufficient to recognize a specific antigen and have specific affinity, and the amino acid residue of antibody residue number 112 by the Chothia method is cysteine. There is something.
The position of the cysteine is a position determined by the Chothia method, but if it is the same position as the amino acid residue, methods other than the Chothia method, for example, those defined by the Kabat method and IMGT method are also included.
Specifically, the VH region in the present invention is a framework in which the amino acid residue of antibody residue number 112 determined by the Chothia method in the three CDRs 1 to 3 in the heavy chain of an antibody against the target antigen molecule is cysteine. It contains at least the region (FR) 4, and preferably comprises the amino acid residues of antibody residues No. 112 determined by the three CDRs 1-3, FR1-3, and Chothia method in the heavy chain of the antibody against the target antigen molecule. It contains at least FR4 whose group is cysteine.
Especially, it consists of the amino acid sequence of antibody residue Nos. 1-112 determined by the Chothia method or the amino acid sequence of antibody residues No. 1-113, and the amino acid residue of antibody residue No. 112 is cysteine. More preferred are those consisting of the amino acid sequence of antibody residues Nos. 1-113 determined by the Chothia method, wherein the amino acid residue of antibody residue No. 112 is cysteine.
The CDRs 1 to 3 may be determined according to the Chothia method, the Kabat method, the IMGT method, other methods, or comprehensively considering them.

  VH (112C) -SARAH according to the present invention is a combination of the C-terminus of the VH region and the N-terminus of the SARAH domain. The VH region and the SARAH domain may be bound directly or via a linker sequence. However, since the expression efficiency and refolding efficiency of VH (112C) -SARAH are high, the linker sequence is It is preferable to pass through. The length of the linker sequence is usually 1 to 4 amino acid residues, preferably 2 amino acid residues. The linker sequence is not particularly limited as long as it does not adversely affect the properties of the fragment antibody of the present invention, and examples thereof include Gly-Ser, Gly-Gly, and Ser-Ser.

（Ｘはアミノ酸残基、ｍは０〜４の整数、−［Ｘ］ｍ−はリンカー配列をそれぞれ表す。）The VH (112C) -SARAH of the present invention is schematically shown below.

(X represents an amino acid residue, m represents an integer of 0 to 4, and-[X] m- represents a linker sequence.)

[VL-SARAH (37C)]
The SARAH domain used here is usually 42 to 54, preferably 43 to 49, more preferably 47 to 49, composed of a short helix (h1) on the N-terminal side and a long helix (h2) on the C-terminal side. Domain, particularly preferably 49 amino acid residues (peptides) having the property of forming an antiparallel coiled coil between each h2 and another SARAH domain, and the C-terminal To the thirteenth amino acid residue is cysteine. H1 preferably consists of 5 to 7 amino acids, and h2 preferably consists of 38 to 42 amino acid residues. In particular, when two SARAH domains form an antiparallel coiled coil between h2, it is preferable that the distance between both N-terminals (of two h1) of the two SARAH domains is 35 to 45 mm, and 39 to 41 mm. The SARAH domain is more preferably 40 Å. As such a SARAH domain, the 13th amino acid residue from the C-terminal of the SARAH domain in the VH (112C) -SARAH of the present invention described above is substituted with cysteine. (Mutated).
Specifically, for example, the SARAH domain of human Mammalian sterile 20-like kinase 1 (Mst1), the SARAH domain of human Mammalian sterile 20-like kinase 2 (Mst2), the SARAH domain of Drosophila Hippo SRAH, SARAH domain, human RASSF1 SARAH domain, human WW45 SARAH domain, Drosophila SAVH SARAH domain, etc., and more than 85% of sequence homology with one of these (preferably SARAH domain of human Mst1) Having a sequence homology of preferably 90% or more, more preferably 95% or more. Te, and amino acid residues 13 th from the C-terminus and the like obtained by substituting (mutation) to a cysteine.
More specifically, for example, the one shown by SEQ ID NO: 9 (the 13th amino acid residue from the C-terminal of the SARAH domain of Mst1 is substituted (mutated) with cysteine), the one shown by SEQ ID NO: 10 (Mst2 The 13th amino acid residue from the C-terminal of the SARAH domain of (SEQ. (Mutated)), represented by SEQ ID NO: 12 (in which the 13th amino acid residue from the C-terminal of the RASSF SARAH domain is substituted (mutated) by cysteine), represented by SEQ ID NO: 13 (in RASSF5) The 13th amino acid residue from the C-terminal of the SARAH domain is substituted (mutated) with cysteine , Represented by SEQ ID NO: 14 (where the 13th amino acid residue from the C-terminal of the RASSF1 SARAH domain is substituted (mutated) with cysteine), or represented by SEQ ID NO: 15 (from the C-terminal of the SAR45 domain of WW45) 13th amino acid residue is substituted (mutated) with cysteine), represented by SEQ ID NO: 16 (13th amino acid residue from the C-terminal of SAV SARAH domain is substituted (mutated) with cysteine), etc. Are preferred, those represented by SEQ ID NOs: 9 to 10 are more preferred, and those represented by SEQ ID NO: 9 are particularly preferred. In addition, as long as it has said property, what is shown by sequence number 9-16 may be a thing by which the 1 or several amino acid in the sequence | arrangement was deleted, substituted, or added. Here, several represents a natural number of 5 or less, preferably 3 or less, more preferably 2 or less. As such, for example, in SEQ ID NOs: 9 to 16, the 26th amino acid residue from the C-terminus is substituted (mutated) with cysteine.

Moreover, even if the SARAH domain in VL-SARAH (37C) according to the present invention is derived from the same SARAH domain as the SARAH domain in VH (112C) -SARAH according to the present invention, it is derived from a different SARAH domain As long as it is easy to form an antiparallel coiled coil, the identity is not limited.
A specific preferred combination of the SARAH domain in the VH (112C) -SARAH according to the present invention and the SARAH domain in the VL-SARAH (37C) according to the present invention is shown by SEQ ID NO: 1 and SEQ ID NO: 13. A combination of those shown in SEQ ID NO: 5 and a combination shown in SEQ ID NO: 9, a combination of those shown in SEQ ID NO: 1 and those shown in SEQ ID NO: 9, and a combination shown in SEQ ID NO: 2. A combination of the one shown in SEQ ID NO: 10, a combination of the one shown in SEQ ID NO: 3 and the one shown in SEQ ID NO: 11, a combination of the one shown in SEQ ID NO: 4 and the one shown in SEQ ID NO: 12. A combination of the one shown in SEQ ID NO: 5 and the one shown in SEQ ID NO: 13, And the combination shown in SEQ ID NO: 14, the combination shown in SEQ ID NO: 7 and the combination shown in SEQ ID NO: 15, and the combination shown in SEQ ID NO: 8 and the combination shown in SEQ ID NO: 16. Preferably, the combination shown by SEQ ID NO: 1 and SEQ ID NO: 9, the combination shown by SEQ ID NO: 2 and the one shown by SEQ ID NO: 10, the combination shown by SEQ ID NO: 1 and SEQ ID NO: More preferred is a combination of the one shown in FIG. 13, the one shown in SEQ ID NO: 5 and the one shown in SEQ ID NO: 9. Those represented by SEQ ID NOs: 1 to 16 may be those in which one or several amino acids in the sequence are deleted, substituted or added. Here, several represents a natural number of 5 or less, preferably 3 or less, more preferably 2 or less.
The SARAH domain in VL-SARAH (37C) according to the present invention is an artificially designed SARAH domain derived from a naturally occurring protein, even if it is a SARAH domain derived from a naturally occurring protein. It may be either.
A SARAH domain derived from a naturally occurring protein is usually used regardless of the species having the protein. However, when the fragment antibody of the present invention is used as a medicine, it is administered in actual use from the viewpoint of antigenicity. Those derived from the same species as the animal to be used are preferred.

The VL region in the present invention is sufficient if it contains at least a site sufficient to recognize an antigen in a specific region and have a specific affinity, for example, 3 in the light chain of an antibody against the target antigen molecule. One containing at least one CDR1 to 3, preferably one containing at least three CDR1 to 1-3 and FR1 to 4 in the light chain of an antibody against the target antigen molecule.
Especially, what consists of an amino acid sequence of the antibody residue numbers 1-107 determined by Chothia method is more preferable.
The CDRs 1 to 3 may be determined according to the Chothia method, the Kabat method, the IMGT method, other methods, or comprehensively considering them.

  The VL-SARAH (37C) of the present invention is a combination of the C-terminal of the VL region and the N-terminal of the SARAH domain. The VL region and the SARAH domain may be bound directly or via a linker sequence. However, since the expression efficiency of VL-SARAH (37C) and the efficiency of refolding are high, the linker sequence is It is more preferable that it couple | bonds through. The length of the linker sequence is usually 1 to 4 amino acid residues, preferably 2 amino acid residues. The linker sequence is not particularly limited as long as it does not adversely affect the properties of the fragment antibody of the present invention, and examples thereof include Gly-Ser, Gly-Gly, and Ser-Ser.

（Ｙはアミノ酸残基、ｌは０〜４の整数、−［Ｙ］ｌ−はリンカー配列をそれぞれ表す。）The VL-SARAH (37C) according to the present invention is schematically shown below.

(Y represents an amino acid residue, l represents an integer of 0 to 4, and-[Y] l- represents a linker sequence.)

[Fragment antibody]
The fragment antibody of the present invention comprises a complex of VH (112C) -SARAH according to the present invention and VL-SARAH (37C) according to the present invention.
More specifically, h2 of the SARAH domain in the VH (112C) -SARAH according to the present invention and h2 of the SARAH domain in the VL-SARAH (37C) according to the present invention are combined by forming an antiparallel coiled coil. Furthermore, the cysteine residue of antibody residue number 112 by the Chothia method of the VH region in VH (112C) -SARAH and the 13th cysteine residue from the C-terminal of the SARAH domain in VL-SARAH (37C) They are linked by disulfide bonds.
Since the fragment antibody of the present invention is bound by such a special binding mode, it not only has antigen-binding activity, but also has higher heat resistance and stability than Fv-class (v1), and Even those obtained in an E. coli expression system have higher crystallization ability of themselves and complexes with antigen molecules than Fv-clasp (v1). As a result, the fragment antibody of the present invention can easily perform a three-dimensional structure analysis of an antigen determination site, which has been difficult with conventional fragment antibodies and antibodies.
Furthermore, since the fragment antibody of the present invention is bound by the special binding mode, it has a function of promoting crystallization of a complex with a protein, particularly a difficult-to-crystallize protein, and promotes protein crystallization. It is particularly useful as a fragment antibody.
In the present specification, “high crystallization ability” means that, when the fragment antibody of the present invention is used to crystallize itself or a complex with an antigen molecule, a high-resolution crystal is obtained. Or / and means that more crystals are obtained in the so-called screening stage.

（図中、Ｘ、Ｙ、ｌ、及びｍは前記と同じ）The fragment antibody of the present invention is schematically shown below.

(In the figure, X, Y, l and m are the same as above)

[Method for Preparing Fragment Antibody of the Present Invention]
The fragment antibody of the present invention can be produced by a general chemical production method according to its amino acid sequence. For example, the fragment antibody of the present invention can be obtained by an ordinary chemical production method (chemical synthesis method) such as the fluorenylmethyloxycarbonyl method (Fmoc method) and the t-butyloxycarbonyl method (tBoc method). It can also be chemically synthesized using a commercially available peptide synthesizer.

  Furthermore, the fragment antibody of the present invention incorporates a nucleic acid molecule encoding the fragment antibody of the present invention into an appropriate expression vector such as a plasmid or phage, and transforms (or transduces) the host cell using the recombinant expression vector. Alternatively, the obtained host cell can be amplified and secreted into or out of the cell by a well-known method using a gene recombination technique.

When preparing the fragment antibody of the present invention by gene recombination technology, for example, a method of preparing according to the following steps may be mentioned.
(1) Gene recombination / expression step (2) Purification step If necessary, a solubilization step, refolding step, concentration step, etc. may be performed after the gene recombination / expression step.

[(1) Gene recombination / expression process]
(1) The gene recombination / expression step is a vector having a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention (hereinafter abbreviated as a VH (112C) -SARAH expression recombinant vector according to the present invention. Or a vector having a nucleic acid sequence encoding the above VL-SARAH (37C) according to the present invention (hereinafter sometimes abbreviated as a VL-SARAH (37C) expression recombinant vector according to the present invention) Is introduced into the host, and the host into which the gene has been introduced is cultured to obtain a crude product containing VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention, or a fragment antibody of the present invention. This is a step of preparing an extract, a culture solution (culture supernatant) or a precipitate.

The nucleic acid sequence encoding VH (112C) -SARAH according to the present invention includes at least the 3 ′ end of the nucleic acid sequence encoding the VH region in the VH (112C) -SARAH and the 5 ′ end of the nucleic acid sequence encoding the SARAH domain. A nucleic acid sequence used for the preparation of a protein in the field of genetic engineering, such as a nucleic acid sequence encoding a start codon, a stop codon, a linker sequence, and a sequence encoding a tag. Etc. may be included.
The nucleic acid sequence encoding the VH region in the VH (112C) -SARAH and the nucleic acid sequence encoding the SARAH domain may be bound directly or via a linker sequence. ) -SARAH expression efficiency and refolding efficiency are high, and therefore, it is preferable to bind via a linker sequence.
The length of the nucleic acid sequence encoding the linker sequence is usually 3 to 12 bases, preferably 6 bases. The nucleic acid sequence encoding the linker sequence is not particularly limited as long as it does not adversely affect the properties of VH (112C) -SARAH according to the present invention and the fragment antibody of the present invention. For example, it encodes Gly-Ser. Examples thereof include a nucleic acid sequence, a nucleic acid sequence encoding Gly-Gly, a nucleic acid sequence encoding Ser-Ser, and the like, and a nucleic acid sequence encoding Gly-Ser is preferred.

The nucleic acid sequence encoding VL-SARAH (37C) according to the present invention includes at least the 3 ′ end of the nucleic acid sequence encoding the VL region and the 5 ′ end of the nucleic acid sequence encoding the SARAH domain in the VL-SARAH (37C). A nucleic acid sequence used for the preparation of a protein in the field of genetic engineering, such as a nucleic acid sequence encoding a start codon, a stop codon, a linker sequence, and a sequence encoding a tag. Etc. may be included.
The nucleic acid sequence encoding the VL region and the nucleic acid sequence encoding the SARAH domain in the VL-SARAH (37C) may be bound directly or via a linker sequence. Since the expression efficiency of SARAH (37C) and the efficiency of refolding are high, it is preferable to bind via a linker sequence.
The length of the nucleic acid sequence encoding the linker sequence is usually 3 to 12 bases, preferably 6 bases. The nucleic acid sequence encoding the linker sequence is not particularly limited as long as it does not adversely affect the properties of VL-SARAH (37C) according to the present invention and the fragment antibody of the present invention. For example, it encodes Gly-Ser. Examples include a nucleic acid sequence, a nucleic acid sequence encoding Gly-Gly, a nucleic acid sequence encoding Ser-Ser, and the like, and a nucleic acid sequence encoding Gly-Ser is preferred.

  The VH (112C) -SARAH expression recombinant vector according to the present invention or the VL-SARAH (37C) expression recombinant vector according to the present invention (hereinafter, the two expression recombinant vectors are combined to form the present invention). May be abbreviated as a recombinant vector for expression according to the present invention) at least a vector having a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention, or VL-SARAH (37C) according to the present invention. A vector having a nucleic acid sequence is not particularly limited as long as it has a function of expressing and producing VH (112C) -SARAH according to the present invention or VL-SARAH (37C) according to the present invention in a host. .

Examples of the recombinant vector for VH (112C) -SARAH expression according to the present invention include SEQ ID NO: 41, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49 according to a conventional cloning technique. , SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52 and other nucleic acid sequences encoding VH (112C) -SARAH according to the present invention, for example, pET16b vector (manufactured by Novagen), pcDNA3.1 vector (Thermo Fisher Scientific) And a vector incorporated into an expression vector such as pCAG-Neo vector (manufactured by Wako Pure Chemical Industries, Ltd.).
Examples of the recombinant vector for expression of VL-SARAH (37C) according to the present invention include, for example, according to a conventional cloning technique, SEQ ID NO: 42, SEQ ID NO: 53, SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57. , SEQ ID NO: 58, SEQ ID NO: 59, SEQ ID NO: 60 and other nucleic acid sequences encoding VH (112C) -SARAH according to the present invention, for example, pET16b vector (manufactured by Novagen), pcDNA3.1 vector (Thermo Fisher Scientific) And a vector incorporated into an expression vector such as pCAG-Neo vector (manufactured by Wako Pure Chemical Industries, Ltd.).

As a method for preparing a recombinant vector for expression according to the present invention, finally, a vector having a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention or VL-SARAH (37C) according to the present invention. Any method can be used as long as it is a known method relating to a method for preparing a recombinant vector for expression by gene recombination, and is not particularly limited.
For example, a method of introducing a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention or a nucleic acid sequence encoding VL-SARAH (37C) according to the present invention into an expression vector can be mentioned.
The nucleic acid sequence encoding the VH (112C) -SARAH according to the present invention or the nucleic acid sequence encoding VL-SARAH (37C) according to the present invention is obtained by introducing the entire length of the nucleic acid sequence into an expression vector at once. Alternatively, gene transfer may be performed in multiple steps.
That is, the recombinant vector for expression according to the present invention is a partial sequence of a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention or a nucleic acid sequence encoding VL-SARAH (37C) according to the present invention. Recombinant expression vector into which the partial sequence has been introduced, the remaining partial sequence of the nucleic acid sequence encoding VH (112C) -SARAH according to the present invention, or the nucleic acid sequence encoding VL-SARAH (37C) according to the present invention May be obtained by introducing the remaining partial sequences.

  A partial sequence of a nucleic acid sequence encoding VH (112C) -SARAH according to the present invention, or a partial sequence of a nucleic acid sequence encoding VL-SARAH (37C) according to the present invention, and VH (112C) -SARAH according to the present invention The remaining partial sequence of the nucleic acid sequence encoding VL-SARAH (37C) according to the present invention is the final partial sequence of VH (112C) -SARAH or the present invention according to the present invention. If a vector having the function of expressing and producing VL-SARAH (37C) according to the invention is obtained, the nucleic acid sequence encoding the original VH (112C) -SARAH according to the present invention, or the VL- It may be separated at any position of the nucleic acid sequence encoding SARAH (37C).

The combination of the partial sequence of the nucleic acid sequence encoding VH (112C) -SARAH according to the present invention and the remaining partial sequence of the nucleic acid sequence encoding VH (112C) -SARAH according to the present invention includes, for example, [the present invention A combination of a nucleic acid sequence encoding the VH region in VH (112C) -SARAH according to the present invention and a sequence encoding the SARAH domain in VH (112C) -SARAH according to the present invention, [VH (112C) according to the present invention] -A combination of a sequence encoding the VH region and linker in SARAH and a nucleic acid sequence encoding the SARAH domain in VH (112C) -SARAH according to the present invention], [VH (112C) -SARAH in the present invention] Nucleic acid sequences encoding VH regions and SARA in VH (112C) -SARAH according to the invention Combinations of sequences] encoding domains and the linker can be mentioned.
The combination of the partial sequence of the nucleic acid sequence encoding VL-SARAH (37C) according to the present invention and the remaining partial sequence of the nucleic acid sequence encoding VL-SARAH (37C) according to the present invention includes, for example, [the present invention A combination of a nucleic acid sequence encoding a VL region in VL-SARAH (37C) according to the present invention and a nucleic acid sequence encoding a SARAH domain in VL-SARAH (37C) according to the present invention, [VL-SARAH according to the present invention ( 37C) in combination with a nucleic acid sequence encoding a VL region in VL-SARAH (37C) according to the present invention and a nucleic acid sequence encoding a SARAH domain and linker in VL-SARAH (37C) according to the present invention, Nucleic acid sequence encoding VL region and linker and SARAH in VL-SARAH (37C) according to the present invention Combination etc. of the nucleic acid sequence encoding the main and the like.
That is, in the preparation of the recombinant expression vector according to the present invention, a partial sequence of the nucleic acid sequence encoding VH (112C) -SARAH according to the present invention, or the nucleic acid encoding VL-SARAH (37C) according to the present invention. A vector having a partial sequence of the sequence (hereinafter sometimes abbreviated as an intermediate vector) may be obtained, for example, a vector having a nucleic acid sequence encoding the SARAH domain in VH (112C) -SARAH according to the present invention. A vector having a nucleic acid sequence encoding a SARAH domain and a linker in VH (112C) -SARAH according to the present invention, a vector having a nucleic acid sequence encoding a SARAH domain in VL-SARAH (37C) according to the present invention, The SARAH domain and linker in VL-SARAH (37C) according to the invention are Vector, and the like having a nucleic acid encoding sequences. These intermediate vectors may contain a restriction enzyme sequence such as NdeI and NcoI upstream of the nucleic acid sequence encoding the linker or the nucleic acid sequence encoding the SARAH domain in VH (112C) -SARAH according to the present invention. .

  Examples of expression vectors include plasmid vectors and virus vectors. Specifically, for example, pET16b vector (manufactured by Novagen), pET28b vector (manufactured by Novagen), pcDNA3.1 vector (manufactured by Thermo Fisher Scientific), pCAG-Neo vector (manufactured by Wako Pure Chemical Industries, Ltd.) ), PTrcHis2 vector, pcDNA3.1 / myc-His vector (Invitrogen), pUC119 (Takara Shuzo), Pqe-tri (Qiagen), pET, pGEX, pMAL plasmid vectors, pMEI-5 DNA ( Virus vectors such as Takara Shuzo Co., Ltd.) and pLVSI-CMV NEO Vector (Takara Shuzo Co., Ltd.). In addition to plasmids derived from E. coli (eg, pTrc99A, pKK223, pET3a), pA1-11, pXT1, pRc / CMV, pRc / RSV, pcDNA I / Neo, p3 × FLAG-CMV-14, pCAT3, pcDNA3.1, Examples thereof include pCMV, and among them, pET16b vector (manufactured by Novagen), pET28b vector (manufactured by Novagen), and pcDNA3.1 vector (manufactured by Thermo Fisher Scientific) are preferable.

For example, when preparing a recombinant vector for VH (112C) -SARAH expression according to the present invention, the SARAH domain in the nucleic acid sequence encoding the VH (112C) -SARAH according to the present invention is added to the expression vector as described above. The intermediate nucleic acid sequence is obtained by introducing a nucleic acid sequence encoding a linker, if necessary, upstream of the nucleic acid sequence encoding the SARAH domain. In the obtained intermediate vector, a VH region in the nucleic acid sequence encoding VH (112C) -SARAH according to the present invention (for example, upstream of the nucleic acid sequence encoding the introduced SARAH domain or the nucleic acid sequence encoding the linker (for example, A nucleic acid sequence encoding three CDRs 1-3, FR1-3, and FR4) in which the amino acid residue of antibody residue number 112 determined by the Chothia method is cysteine in the heavy chain of an antibody against the target antigen molecule By introducing, recombinant vectors for VH (112C) -SARAH expression against various antigen molecules can be easily prepared.
More specifically, if an expression vector (intermediate vector) as described above into which a linker nucleic acid sequence (for example, Gly-Ser) and a nucleic acid sequence encoding a SARAH domain have been prepared, the linker nucleic acid sequence of the expression vector is prepared. And upstream of the nucleic acid sequence encoding the SARAH domain, a desired VH region portion (for example, antibody residues determined by the three CDR1 to 3 and FR1 to 3 and Chothia method in the heavy chain of the antibody against the antigen molecule of interest) Introducing FR4 in which the amino acid residue of No. 112 is cysteine is included), recombinant vectors for VH (112C) -SARAH expression against various antigen molecules can be easily prepared.
The VL-SARAH (37C) expression recombinant vector according to the present invention can be prepared in the same manner as the VH (112C) -SARAH expression recombinant vector according to the present invention.

  VH (112C) -SARAH according to the present invention and / or VL-SARAH (37C) according to the present invention may be expressed as a fusion protein with a tag peptide or another protein. Examples of the tag peptide to be fused include a FLAG tag, a 3X FLAG tag, a His tag (His tag, for example, 6 × His tag) and the like.

Subsequently, a transformant is prepared by transforming (transducing) an appropriate host cell using the recombinant vector for expression.
Any host cell may be used as long as it has a function of expressing and producing VH (112C) -SARAH according to the present invention and / or VL-SARAH (37C) according to the present invention. For example, Escherichia and prokaryotic organisms such as B. subtilis, B. brevis, B. borstelenis, animal cells, insect cells, plant cells, yeast cells and the like. Mammalian cells are more preferred, and E. coli is particularly preferred. Examples of mammalian cells include Expi293F cells, HEK293T cells, COS-7 cells, CHO-K1 cells, CHO-S cells, and the like. Examples of E. coli include BL21 (DE3), K-12, DH1, DH5, DH5α, M15, HB101, C600, XL-1 Blue, JM109, JM105, JM127, XL1-Blue, VCS257, and TOP10.
Moreover, you may use a competent cell (Competent Cell) with higher introduction efficiency of plasmid or phage DNA. Examples of competent cells include E.I. E. coli DH5α Competent Cell, E. coli. E. coli JM109 Competent Cells (manufactured by Takara Bio Inc.) and the like.

As described above, Fv-class (v1) obtained by using E. coli as a host cell (hereinafter sometimes abbreviated as E. coli expression system) does not have the ability to crystallize a complex with an antigen molecule. Even if it has the crystallization ability, only low-resolution crystals can be obtained at the so-called screening stage before the optimization of the crystallization conditions.
On the other hand, even when the fragment antibody of the present invention is prepared in an E. coli expression system, one having the ability to crystallize the fragment antibody of the present invention itself or a complex with a specific antigen can be obtained.

The VH (112C) -SARAH expression recombinant vector according to the present invention and the VL-SARAH (37C) expression recombinant vector according to the present invention may be co-expressed or expressed in separate host cells. However, when E. coli is used as a host cell, it is preferably expressed in a separate host cell, and when an animal cell is used as the host cell, it is preferably co-expressed.
When co-expressed in animal cells, it will be described later because it is expressed as a fragment antibody of the present invention (a complex of VH (112C) -SARAH according to the present invention and VL-SARAH (37C) according to the present invention). A refolding step is not required. However, when expressing in different host cells, the expressed VH (112C) -SARAH according to the present invention and VL-SARAH (37C) according to the present invention are mixed, and the refolding step described later is performed. It is necessary to perform a folding process.

  Examples of a method for introducing a recombinant vector for expressing VH (112C) -SARAH according to the present invention or a recombinant vector for expressing VL-SARAH (37C) according to the present invention into a host cell include, for example, “proteins that can be selected according to purpose” What is necessary is just to follow according to the well-known method as described in an expression protocol, a 3rd chapter protein expression protocol, ISBN978-4-7581-0175-2, sheep soil company.

The transformant is then cultured in a nutrient medium to produce VH (112C) -SARAH according to the invention or VL-SARAH (37C) according to the invention. The culture is performed by a known method, and the temperature, the pH of the medium, and the culture time may be set as appropriate.
When cultivating a transformant whose host cell is Escherichia coli, it may be carried out in a liquid medium that is usually used under the usual conditions for culturing Escherichia coli, and a transformant whose host cell is an animal cell is cultured. In this case, it may be carried out in a liquid medium that is usually used under the usual conditions for culturing animal cells. Further, for example, an agent such as isopropyl-β-D-thiogalactopyranoside (IPTG) or 3β-indolylacrylic acid may be added as necessary.

A culture solution (culture supernatant) crude extract or precipitate containing VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention or a fragment antibody of the present invention is obtained by the above culture. It can be obtained from the culture as follows.
That is, when the VH (112C) -SARAH according to the present invention, the VL-SARAH (37C) according to the present invention or the fragment antibody of the present invention is secreted into the culture medium of the transformant, the present invention A culture solution (culture supernatant) containing VH (112C) -SARAH, VL-SARAH (37C) according to the present invention or the fragment antibody of the present invention is obtained.
When VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention or the fragment antibody of the present invention is present in the periplasm or cytoplasm of the transformant, a method such as filtration or centrifugation To recover cells or cells from the culture and resuspend them in a suitable buffer. Then, after disrupting the cell wall and / or cell membrane of the collected cells by a method such as surfactant treatment, ultrasonic treatment, lysozyme treatment, freeze thawing, etc., the method is applied to the present invention by a method such as centrifugation or filtration. A crude extract containing the VH (112C) -SARAH according to the present invention, the VL-SARAH (37C) according to the present invention or the fragment antibody of the present invention is obtained.
When the VH (112C) -SARAH according to the present invention, the VL-SARAH (37C) according to the present invention, or the fragment antibody of the present invention is expressed as inclusion bodies, the bacteria are removed from the culture by a method such as filtration or centrifugation. Body or cells are collected and resuspended in an appropriate buffer. Then, the cell wall and / or cell membrane of the collected cells and the like are destroyed by a method such as surfactant treatment, ultrasonic treatment, lysozyme treatment, freeze thawing, etc., and then centrifuged according to a method such as centrifugation. A precipitate containing (112C) -SARAH, VL-SARAH (37C) according to the invention or a fragment antibody of the invention is obtained. In this case, it is necessary to perform a solubilization step described later.

  In the method for preparing a fragment antibody of the present invention, a solubilization step, a refolding step, a concentration step, and the like are performed as necessary after the above (1) gene recombination / expression step.

[Solubilization process]
The solubilization step was the above-described (1) gene recombination / expression step, in which VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention or the fragment antibody of the present invention was expressed as inclusion bodies. In some cases, the recovered VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention, or the precipitate containing the fragment antibody of the present invention, VH (112C) -SARAH according to the present invention, The VL-SARAH (37C) or the fragment antibody of the present invention is subjected to a solubilization method to obtain a solubilized solution containing them.

The solubilization method may be a known method and is not particularly limited.
Specifically, the precipitate containing the VH (112C) -SARAH according to the present invention, the VL-SARAH (37C) according to the present invention, or the fragment antibody of the present invention recovered in the above (1) gene recombination / expression step. A solubilized solution may be added to obtain a solubilized solution containing VH (112C) -SARAH according to the present invention, VL-SARAH (37C) according to the present invention, or a fragment antibody of the present invention.

  The solubilization solution is not particularly limited as long as it can solubilize the precipitate of the inclusion body obtained in step (1), and examples thereof include known solubilization solutions containing 6M guanidine hydrochloride, 8M urea and the like. More specifically, 6M guanidine hydrochloride solubilized solution [6M guanidine hydrochloride, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl], 8M urea solubilized solution [8 M Urea, 50 mM Tris-HCl (pH 8) 0.0), 150 mM NaCl] and the like.

[Refolding process]
The refolding step is performed when the VH (112C) -SARAH according to the present invention and the VL-SARAH (37C) according to the present invention are expressed in different hosts. (112C) -SARAH solubilized solution, VH (112C) -SARAH crude extract obtained in the above (1) gene recombination / expression step, or (1) gene recombination / expression step VH (112C) -SARAH according to the present invention in the culture solution (culture supernatant) of VH (112C) -SARAH according to the present invention obtained in
Solubilized solution of VL-SARAH (37C) according to the present invention obtained in the above-mentioned solubilization step, rough extraction of VL-SARAH (37C) according to the present invention obtained in the above (1) gene recombination / expression step Or VL-SARAH (37C) according to the present invention in the culture solution (culture supernatant) of VL-SARAH (37C) according to the present invention obtained in the above (1) gene recombination / expression step This is a method of forming a body by a refolding method.

The refolding method is not particularly limited as long as it can form a complex of VH (112C) -SARAH according to the present invention and VL-SARAH (37C) according to the present invention.
Specifically, the solubilized solution of VH (112C) -SARAH according to the present invention obtained in the above-described solubilization step, and the VH (112C) according to the present invention obtained in the above-mentioned (1) gene recombination / expression step. -A crude extract of SARAH, or a culture solution (culture supernatant) of VH (112C) -SARAH according to the present invention obtained in the above (1) gene recombination / expression step,
A solubilized solution of VL-SARAH (37C) according to the present invention obtained in the solubilization step, a crude extract of VL-SARAH (37C) according to the present invention obtained in the above (1) gene recombination step, Alternatively, the culture solution (culture supernatant) of VL-SARAH (37C) according to the present invention obtained in the above (1) gene recombination / expression step is mixed, and one or a plurality of types of refolding are mixed in the obtained mixture solution A solution is added to form a fragment antibody of the present invention (VH (112C) -SARAH according to the present invention and VL-SARAH (37C) complex according to the present invention), and refolding containing the fragment antibody of the present invention What is necessary is just to obtain a used solution.

  Examples of the refolding solution include the refolding solution A [4M Urea, 0.4M L-Arg, 50 mM Tris-HCl (pH 8.07.5), 150 mM NaCl, 375 μM oxidized glutathione], and the refolding solution B [ 0.4M L-Arg, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 375 μM oxidized glutathione] and the like.

[Concentration process]
The concentration step is a step of concentrating the refolded solution obtained in the refolding step to obtain a concentrated solution. The concentration step is performed, for example, when the protein is diluted by the above refolding step or the like, whereby the subsequent purification step can be easily performed.

  The concentration method is not limited as long as it can concentrate the refolded solution obtained in the above refolding step, and examples thereof include known concentration methods such as salting out method and ultrafiltration method. .

[(2) Purification step]
In the purification step, the fragment antibody of the present invention is isolated from the culture solution (culture supernatant) or crude extract obtained in the above-mentioned (1) gene recombination / expression step, or the concentrate obtained in the concentration step, It is a process to purify.

  Examples of the method for isolating and purifying the fragment antibody of the present invention include a method utilizing charge such as ion exchange chromatography, a method utilizing specific affinity such as affinity chromatography, and reverse phase high performance liquid chromatography. Method utilizing difference in hydrophobicity, method utilizing difference in isoelectric point such as isoelectric focusing, method utilizing solubility such as salting out and solvent precipitation method, method utilizing difference in molecular weight such as gel filtration , Etc.

  The purified fragment antibody of the present invention can be confirmed by, for example, SDS-PAGE. For example, by confirming the molecular weight by SDS-PAGE etc., the VH (112C) -SARAH according to the present invention and the VL-SARAH (37C) according to the present invention form a heterodimer and have a correct three-dimensional structure, , The cysteine residue of antibody residue number 112 by the Chothia method in the VH region of VH (112C) -SARAH according to the present invention and the thirteenth position from the C-terminus of the SARAH domain in VL-SARAH (37C) It can be confirmed whether or not the cysteine residues of each form an SS bond.

[Protein crystallization method]
The protein crystallization method of the present invention is a method of crystallizing a protein (hereinafter sometimes referred to as a target protein) in the presence of the fragment antibody of the present invention. The target protein is a protein (antigen) that is specifically recognized by the fragment antibody of the present invention.

That is, the protein crystallization method of the present invention may be a known crystallization method in the presence of the fragment antibody of the present invention, and is not particularly limited as long as it can obtain a target protein crystal.
Specific examples of the crystallization method include a batch method, a vapor diffusion method, a dialysis method, a free interface diffusion method, a concentration method, a light irradiation method, a vibration method, and a stirring method, and the vapor diffusion method is more preferable. Examples of the vapor diffusion method include a sitting drop method, a hanging drop method, and a sandwich drop method, and the sitting drop method is more preferable.

More specifically, the protein crystallization method of the present invention includes the following steps.
(1) A step of mixing a solution containing the target protein and a solution containing the fragment antibody of the present invention to form a complex of the target protein and the fragment antibody of the present invention (hereinafter abbreviated as complex formation step). May be)
(2) A step of obtaining a crystal of a complex of the target protein and the fragment antibody of the present invention (hereinafter, crystallization step)

[Composite formation step]
The complex formation step is performed by mixing a solution containing the target protein and a solution containing the fragment antibody of the present invention to form a complex of the target protein and the fragment antibody of the present invention.

The solvent containing the target protein is not particularly limited as long as it is generally used for protein crystallization. For example, pure water, buffer solution, precipitant solution, surfactant solution, alcohol solution, organic solvent, chelate An agent solution, a reducing agent solution, a cryoprotectant solution, an ionic liquid, and a mixed solution thereof are exemplified, and pure water and a buffer solution are preferable. Although the pH of a buffer solution is not normally limited, For example, it is pH 4-9, Preferably it is pH 5-8. Specific examples include phosphate buffered saline (PBS), HEPES buffer, borate buffer, Tris buffer, phosphate buffer, veronal buffer, Good buffer, and the like. In addition, the buffering agent concentration of these buffers is appropriately selected from the range of usually 5 to 100 mM, preferably 5 to 20 mM.
The solvent containing the fragment antibody of the present invention is the same as the solvent containing the target protein, and preferred specific examples are also the same.

  The amount of the target protein in the solution containing the target protein and the amount of the fragment antibody of the present invention in the solution containing the fragment antibody of the present invention are determined by mixing the solution containing the target protein and the solution containing the fragment antibody of the present invention. In the solution, the molar equivalent of the target protein is equal to or higher than the molar equivalent of the fragment antibody of the present invention, and the concentration of the complex of the target protein and the fragment antibody of the present invention in the mixed solution is usually 2 mg / mL to An amount of 100 mg / mL, preferably 5 mg / mL to 30 mg / mL, more preferably 10 mg / mL to 20 mg / mL may be used.

  A mixture of the solution containing the target protein and the solution containing the fragment antibody of the present invention (hereinafter sometimes abbreviated as a crystallized sample) is concentrated by a known method such as an ultrafiltration device, if necessary. May be.

[Crystalling process]
The crystallization step is a step of obtaining crystals of a complex of the target protein formed in the complex formation step and the fragment antibody of the present invention by a crystallization method known per se, for example, a specific crystallization method as described above. is there.

Specifically, the crystallization step may be performed as follows, for example.
The mixed solution (crystallized sample) containing the complex of the target protein obtained in the complex formation step and the fragment antibody of the present invention is usually 0.05 μL to 2 μL, preferably 0.1 μL to 0.5 μL per condition. Then, the screening kit solution attached to the screening kit is mixed with the same amount to make a drop, and the reservoir (precipitant solution) having the same composition as the crystallized sample (usually 50 μL to 1000 μL, preferably 80 μL to 150 μL), Crystallization is performed by a sitting drop method to obtain a complex crystal of the target protein and the fragment antibody of the present invention.

  In addition, if necessary, by subjecting crystals of the complex of the target protein obtained by the crystallization step and the fragment antibody of the present invention to a known method such as a step of detecting, a step of analyzing by an X-ray diffractometer, The three-dimensional structure of the complex of the target protein and the fragment antibody of the present invention can be analyzed.

  According to the protein crystallization method of the present invention, high-resolution crystals can be obtained by using the fragment antibody of the present invention, and more crystals can be obtained in the so-called screening stage.

  Even when the fragment antibody of the present invention is prepared in an E. coli expression system that is easier to handle than an animal cell system, it has higher antigen-binding activity, higher crystallization ability than Fv-class (v1), and protein crystallization. Since it has the property to promote, etc., it is useful in the field | area where protein crystallization, such as a drug discovery, is performed as a fragment antibody which has the said property which can be manufactured easily. In addition, the fragment antibody of the present invention has heat resistance and high stability, and thus is useful in the field of antibody medicine.

  EXAMPLES Hereinafter, although this invention is demonstrated concretely based on an Example and a comparative example, this invention is not limited at all by these examples.

<Preparation of Fv-class (v1) of P20.1 antibody using E. coli expression system in Comparative Example 1>
By the following method, VH-SARAH (Y35C) against the P20.1 antibody represented by SEQ ID NO: 17 (hereinafter sometimes abbreviated as P20.1 antibody-VH-SARAH (Y35C)), and SEQ ID NO: 18 VL-SARAH (M24C) (hereinafter sometimes abbreviated as P20.1 antibody-VL-SARAH (M24C)) against the indicated P20.1 antibody (1) Gene recombination / expression step and solubilization step (2) Refolding step, (3) Concentration step, and (4) Purification step, Fv-class (v1) against P20.1 antibody (hereinafter referred to as P20.1 antibody-Fv-clasp (v1)) May be abbreviated).
The obtained P20.1 antibody-Fv-clas (v1) is composed of the cysteine residue 35th from the N-terminus in the SARAH domain of P20.1 antibody-VH-SARAH (Y35C), and P20.1 antibody- It is designed so that the 24th cysteine residue from the N-terminal in the SARAH domain of VL-SARAH (M24C) forms an SS bond.
(1) Gene recombination / expression step and solubilization step (1) -1: Preparation of VH-SARAH (Y35C) against P20.1 antibody P20.1 antibody-VH-SARAH (Y35C) represented by SEQ ID NO: 19 In accordance with a conventional method, a nucleic acid sequence coding for is introduced into the NcoI / HindIII restriction enzyme site of pET28b (+) vector (manufactured by Novagen) to prepare a recombinant vector for expression of P20.1 antibody-VH-SARAH (Y35C) did. The obtained recombinant vector for P20.1 antibody-VH-SARAH (Y35C) expression was introduced into E. coli BL21 (DE3) strain (manufactured by Novagen). E. coli BL21 (DE3) strain into which the recombinant vector for expression of P20.1 antibody-VH-SARAH (Y35C) was introduced was cultured at 37 ° C. using 1 L of LB medium. Induction was performed by adding IPTG to the medium so that the final concentration was about 0.4 mM at around OD = 0.6. Four hours after the start of the culture, E. coli BL21 (DE3) strain (cells) into which the recombinant vector for expression of P20.1 antibody-VH-SARAH (Y35C) was introduced was recovered.
Among the collected cells, the cells obtained by culturing in 0.5 L of the medium are suspended in 20 mL of TBS (20 mL Tris-HCl (pH 7.5), 150 mM NaCl). Obtained. To the obtained suspension, the cells disruption reagents [Peptstatin (final concentration is 1 μM), Leupeptin (final concentration is 10 μM), PMSF (final concentration is 250 μM) and Turbonuclease (manufactured by Accelgen, 3 μL)] are added. The cells were crushed using an ultrasonic generator (TOMY) and then centrifuged (25000 × g, 4 ° C., 20 min), and the supernatant was removed to obtain a precipitate. To the resulting precipitate, 6 mL of solubilized solution [6M guanidine hydrochloride, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl)] was added and incubated at 37 ° C. for 2 hours with occasional pipetting. . Subsequently, centrifugation (25000 × g, 4 ° C., 20 min) was performed, and the supernatant containing P20.1 antibody-VH-SARAH (Y35C) was recovered.
(1) -2: Preparation of VL-SARAH (M24C) against P20.1 antibody The pET16b vector was prepared by using a nucleic acid sequence encoding P20.1 antibody-VL-SARAH (M24C) represented by SEQ ID NO: 20 according to a conventional method. It was introduced into the NcoI / XhoI restriction enzyme site (manufactured by Novagen) to prepare a recombinant vector for expression of P20.1 antibody-VL-SARAH (M24C). The obtained recombinant vector for P20.1 antibody-VL-SARAH (M24C) expression was introduced into E. coli BL21 (DE3) strain (manufactured by Novagen).
After the introduction of the expression vector into Escherichia coli, P20.1 antibody-VL-SARAH (M24C) is prepared by the same method as in (1) -1 above, and P20.1 antibody-VL-SARAH (M24C) is included. The supernatant was collected.
(2) Refolding step The supernatant containing P20.1 antibody-VH-SARAH (Y35C) recovered in (1) -1 above, and the P20.1 antibody-VL-SARAH (recovered in (1) -2 above) The supernatant containing M24C) was mixed to obtain a mixture. The obtained mixed solution was filtered with a 0.45 μm filter, and the refolding solution A [4M Urea, 0.4M L-Arg, 50 mM Tris-HCl (pH 8. 0), 150 mM NaCl, 375 μM oxidized glutathione] was added to the filtrate with stirring and incubated at 4 ° C. for 4 hours. Refolding solution B [0.4 M L-Arg, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 375 μM oxidized glutathione] was added so that the obtained mixed solution after incubation was diluted 2-fold. After adding to the filtrate with stirring, it was incubated at 4 ° C. overnight (overnight) to obtain a refolded antibody solution.
(3) Concentration step Ammonium sulfate ground to 80% saturation was added little by little to the refolded antibody solution obtained in (2) above, followed by ammonium sulfate precipitation, followed by stirring at 4 ° C for 3 hours. Incubated while. The obtained solution after incubation was centrifuged (12000 × g, 4 ° C., 60 min), and the supernatant was removed to obtain a precipitate. A refolding solution [a solution obtained by mixing the above refolding solution A and the above refolding solution B in a ratio of 1: 1] was added to the obtained precipitate and suspended therein, thereby obtaining a suspension. The obtained suspension was dialyzed 3 times in total for 4 to 16 hours using 1 L of dialysis solution [50 mM Tris-HCl (pH 8.0), 150 mM NaCl] according to a conventional method. The suspension after dialysis was collected, and then centrifuged (25000 × g, 4 ° C., 20 min), and the supernatant was collected. The obtained supernatant was concentrated by ultrafiltration according to a conventional method to obtain a concentrated solution.
(4) Purification step The concentrated solution obtained in the above (3) is subjected to HiLoad 16/600 Superdex 200 pg (manufactured by GE) and gel filtration purification elution buffer [50 mM Tris-HCl (pH 8) according to a conventional method. 0.0), 150 mM NaCl], gel filtration purification was performed at a flow rate of 1.0 mL / min. Since a monodispersed peak was obtained at a position estimated as P20.1 antibody-Fv-clas (v1) from the molecular weight marker, the eluate of the peak was collected.
Subsequently, the eluate collected by gel filtration purification was subjected to SDS-PAGE according to a conventional method. As a result, in addition to the band of P20.1 antibody-Fv-clasp (v1), an impurity band was also confirmed.
Therefore, in order to remove impurities from the eluate collected by gel filtration purification, CNBr-activated Sepharose 4 Fast Flow (manufactured by GE) in which C8 peptide (PRGYPGQV), which is an antigen molecule of the fragment antibody represented by SEQ ID NO: 21, is immobilized. ), Equilibration buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl], elution buffer [0.5 mg / mL C8, 20 mM Tris-HCl (pH 7.5), 150 mM NaCl]. The eluate was collected.
The collected eluate was subjected to SDS-PAGE under reducing conditions or non-reducing conditions, respectively, according to a conventional method.
The results are shown in FIG. In the figure, the left lane is an electrophoretic diagram under reducing conditions, and the right lane is an electrophoretic diagram under non-reducing conditions. Under reducing conditions, a band of P20.1 antibody-VH-SARAH (Y35C) was obtained around 18 kDa, and a P20.1 antibody-VL-SARAH (M24C) band was obtained around 16 kDa. Further, a band was obtained around 35 kDa under non-reducing conditions. The P20.1 antibody-VH-SARAH (Y35C) and the P20.1 antibody-VL-SARAH (M24C) have Cys residues so that an SS bond is formed only when they both correctly form heterodimers. Since the introduction position was designed, these results showed that P20.1 antibody-Fv-clas (v1), a heterodimer in which the S—S bond between chains was correctly formed, was obtained.

<Comparative Example 2 Preparation of Fv-class (v1) of P20.1 Antibody Using Animal Cell Expression System>
According to the following method, VH-SARAH (Y35C) against the P20.1 antibody represented by SEQ ID NO: 22 (containing a mouse knight signal sequence at the N-terminus) and VL- against the P20.1 antibody represented by SEQ ID NO: 24 SARAH (M24C) (containing a mouse nitrogen signal sequence at the N-terminus) was used for (1) gene recombination step, (2) recombinant protein expression step using animal cell expression system, and (3) purification step To obtain Fv-class (v1) of the P20.1 antibody (hereinafter sometimes abbreviated as P20.1 antibody-Fv-class (v1)).
(1) Gene recombination step A nucleic acid sequence encoding VH-SARAH (Y35C) against the P20.1 antibody represented by SEQ ID NO: 23 (including a sequence encoding a mouse nitrogen signal sequence at the N-terminus) Then, it was introduced into the HindIII / PmeI restriction enzyme site of pcDNA3.1 vector (manufactured by Thermo Fisher Scientific) to prepare a recombinant vector for expression of P20.1 antibody-VH-SARAH (Y35C). Further, in place of the nucleic acid sequence shown in SEQ ID NO: 23, a nucleic acid sequence encoding VL-SARAH (M24C) against the P20.1 antibody shown in SEQ ID NO: 25 (a sequence encoding a mouse nitrogen signal sequence at the N-terminus) Was used to prepare a recombinant vector for expression of P20.1 antibody-VL-SARAH (M24C).
(2) Recombinant protein expression step using animal cell expression system Expi293F cells (manufactured by Thermo Fisher Scientific) are used in 600 mL of Expi293 Expression Medium (manufactured by Thermo Fisher Scientific), and the cells are 3 × 10 ^6. seeded to cells / mL. Thereafter, in accordance with a conventional method, 400 μg of each of the P20.1 antibody-VH-SARAH (Y35C) expression recombinant vector and the P20.1 antibody-VL-SARAH (Expirectamine 293 Reagent (manufactured by Thermo Fisher Scientific)) M24C) Recombinant expression vector was transfected into Expi293F cells. After the gene introduction, Expi293F cells transfected with 125 rpm at 37 ° C. and 8% CO 2 were cultured with shaking for 18 hours. Thereafter, 3 mL and 30 mL of Expifectamine 293 Transfection Enhancer 1 and Expifectamine 293 Transfection Enhancer 2 (manufactured by Thermo Fisher Scientific) were added, respectively, and the culture was shaken at 125 rpm and 8% CO2 for 3 days. Was recovered.
(3) Purification step The culture supernatant collected in (2) was purified using CNBr-activated Sepharose 4 Fast Flow (GE) in which C8 peptide was immobilized in the same manner as in Comparative Example 1. Subsequently, the elution fraction was added with Superdex 200 Iclear 10/300 GL (manufactured by GE) and gel filtration purification elution buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl] according to a conventional method. And gel filtration purification was performed at a flow rate of 0.5 mL / min. Since a monodispersed peak was obtained at a position estimated as P20.1 antibody-Fv-clas (v1) from the molecular weight marker, the eluate of the peak was collected.
The collected eluate was subjected to SDS-PAGE under reducing conditions or non-reducing conditions, respectively, according to a conventional method.
The results are shown in FIG. In the figure, the left lane is an electrophoretic diagram under reducing conditions, and the right lane is an electrophoretic diagram under non-reducing conditions. Under reducing conditions, a band of P20.1 antibody-VH-SARAH (Y35C) was obtained around 18 kDa, and a P20.1 antibody-VL-SARAH (M24C) band was obtained around 16 kDa. Further, a band was obtained around 35 kDa under non-reducing conditions. The P20.1 antibody-VH-SARAH (Y35C) and the P20.1 antibody-VL-SARAH (M24C) have Cys residues so that an SS bond is formed only when they both correctly form heterodimers. Since the introduction position was designed, these results showed that P20.1 antibody-Fv-clas (v1), a heterodimer in which the S—S bond between chains was correctly formed, was obtained.

<Comparative Example 3 Preparation of Fv-class (v1) of 12CA5 Antibody Using E. coli Expression System>
Instead of the nucleic acid sequence encoding P20.1 antibody-VH-SARAH (Y35C) shown in SEQ ID NO: 19, a nucleic acid sequence encoding VH-SARAH (Y35C) against 12CA5 antibody shown in SEQ ID NO: 26 was used. In addition, in place of the nucleic acid sequence encoding VL-SARAH (M24C) for the P20.1 antibody represented by SEQ ID NO: 20, the nucleic acid sequence encoding VL-SARAH (M24C) for the 12CA5 antibody represented by SEQ ID NO: 27 Except for using VH-SARAH (Y35C) against the 12CA5 antibody represented by SEQ ID NO: 28 (hereinafter sometimes abbreviated as 12CA5 antibody-VH-SARAH (Y35C)) in the same manner as in Comparative Example 1 And VL-SARA against 12CA5 antibody shown in SEQ ID NO: 29 (M24C) (hereinafter sometimes abbreviated as 12CA5 antibody-VL-SARAH (M24C)), (1) gene recombination / expression step and solubilization step, (2) refolding step, and (3) concentration The process was followed by (4) purification step to obtain 12CA5 antibody Fv-class (v1) (hereinafter sometimes abbreviated as 12CA5 antibody-Fv-class (v1)).
The results of (4) purification step are shown below.
The concentrated solution obtained in the concentration step (3) was subjected to gel filtration purification in the same manner as in Comparative Example 1. As a result, the monodisperse was estimated at the position estimated as 12CA5 antibody-Fv-class (v1) from the molecular weight marker. Since the peak was obtained, the eluate of the peak was collected.
Subsequently, the recovered eluate was subjected to non-reducing SDS-PAGE according to a conventional method. As a result, in addition to the 12CA5 antibody-Fv-clasp (v1) band near 37 kDa, an impurity band was also confirmed near 15-18 kDa. 12CA5 antibody-VH-SARAH (Y35C) and 12CA5 antibody-VL-SARAH (M24C) are designed to introduce Cys residues so that SS bonds are formed only when they form a heterodimer correctly. Therefore, these impurity bands were considered to be the respective homodimers or unoxidized heterodimers due to partially incomplete refolding.
Therefore, in order to remove impurities from the eluate recovered by gel filtration purification, according to a conventional method, Mono Q 5/50 (manufactured by GE), elution buffer for ion exchange chromatography [50 mM to 300 mM gradient of NaCl in Tris- Using HCl, pH 7.5], ion exchange chromatography was performed by applying a gradient of 20 mL at a flow rate of 1 mL / min.
As a result, after the largest peak (hereinafter sometimes abbreviated as peak A), two small peaks (hereinafter sometimes abbreviated as peaks B and C) were obtained.
The collected eluates of peaks A to C were subjected to SDS-PAGE under non-reducing conditions and reducing conditions, respectively, according to a conventional method.
The results are shown in FIG. In the figure, the left gel is an electrophoresis diagram under reducing conditions, and the right gel is an electrophoresis diagram under non-reducing conditions. The lane names A to C correspond to the names of the eluates of peaks A to C collected by gel filtration purification, respectively.
As a result of electrophoresis under non-reducing conditions, A is a band of 12CA5 antibody-Fv-class (v1) in the vicinity of 35 kDa, B is a band of 12CA5 antibody-VH-SARAH (Y35C) in the vicinity of 18 kDa and 12CA5 antibody in the vicinity of 16 kDa A band of -VL-SARAH (M24C) and a band of 12CA5 antibody-VL-SARAH (M24C) were obtained in the vicinity of 16 kDa, respectively.
From these results, it was found that 12CA5 antibody-Fv-clas (v1), which is a heterodimer in which an S—S bond between chains was correctly formed, was obtained in peak A. Peak B was considered as an unoxidized heterodimer molecule, and peak C was considered as a homodimer molecule.

<Comparative Example 4 Crystal Structure Analysis of Complex of Fv-class (v1) of P20.1 Antibody and C8 Peptide Obtained in E. coli Expression System>
By using the P20.1 antibody-Fv-class (v1) obtained by the E. coli expression system in Comparative Example 1 by the following method, the P20.1 antibody-Fv-class (v1) and the antigen molecule of the fragment antibody were A complex with a C8 peptide was crystallized. Both were mixed so that the final concentrations of P20.1 antibody-Fv-clasp (v1) and C8 peptide would be 7 mg / mL and 1 mM, respectively, to obtain a crystallized sample. Automatic crystallizer mosuquito (manufactured by TTP LabTech), various screening kits (Classics Neo Suite (manufactured by QIAGEN) (96 conditions), Wizard Classic 1 & 2 (manufactured by Rigaku) (96 conditions), Wizard Precipitang (192 conditions), SaltRx (manufactured by Hampton Research) (96 conditions)], violamoprotein crystallization plate (96 well, manufactured by ASONE), 0.1 μL of crystallization sample per condition and 0 .1 μL of screening kit solution (crystallization reagent) was mixed to prepare a drop, and 80 μL of crystallization reagent was used as a reservoir, and crystallization was screened by a sitting drop vapor diffusion method.
As described above, a total of 480 screening conditions were examined using four types of kits, but no crystals could be obtained. It was found that the Fv-clasp (v1) obtained by the E. coli expression system does not have a crystallization ability in the complex of the fragment antibody and the antigen molecule.

<Comparative Example 5 Crystal Structure Analysis of Complex of Fv-Clasp (v1) of P20.1 Antibody and C8 Peptide Obtained in Animal Cell Expression System>
Using the P20.1 antibody-Fv-class (v1) obtained by the animal cell expression system in Comparative Example 2, the P20.1 antibody-Fv-class (v1) and the fragment thereof were produced in the same manner as in Comparative Example 4. Crystallization of the complex with the C8 peptide, which is the antigen molecule of the antibody, was performed. However, Index (manufactured by Hampton Research) and Wizard Classic 1 & 2 (manufactured by Rigaku) were used as the screening kit.
As a result, crystals of the complex of P20.1 antibody-Fv-class (v1) and C8 peptide in the screening stage were obtained under 7 conditions (3.6%) out of 192 conditions. A crystal (Crystal-1) obtained by further optimizing the condition for one of these conditions, and a crystal (Crystal-2) obtained under the conditions different from Crystal-1 in the screening, the synchrotron radiation facility SPring- An X-ray diffraction experiment was conducted at 8. As a result, X-ray diffraction data was obtained with a resolution of 1.31Å for Crystal-1 and a resolution of 1.75Å for Crystal-2. The obtained X-ray diffraction data was used as a search model for the crystal structure of P20.1 Fab (PDB ID: 2ZPK) and the SARAH domain crystal structure of Mst1 (PDB ID: 4NR2). J. et al. Et al. Phaser crystallographic software. J. et al. Appl. Crystallogr. 40, 658-674 (2007)], and the phase was determined. Then, REFMAC5 [Murshudov, G. N. Et al. REFMAC 5 for the refinement of macromolecular crystal structures. Acta Crystallogr. D Biol. Crystallogr. 67, 355-367 (2011)] and PHENIX [Adams, P. et al. Et al. Reent developments in the PHENIX software for automated crystallographic structure determination. J. et al. Synchrotron. Radiat. 11, 53-55. (2004)] was used to refine the structure. FIG. 4 shows the results of X-ray crystal structure analysis of the complex (Crystal-1) of the obtained P20.1 antibody Fv-class (v1) and C8 peptide.

<Comparative Example 6 Crystal Structure Analysis of Complex of 12 CA5 Antibody Fv-Clasp (v1) and HA Peptide Obtained in E. coli Expression System>
By using the 12CA5 antibody-Fv-clas (v1) obtained by the E. coli expression system in Comparative Example 3 in the same manner as in Comparative Example 4, 12CA5 antibody-Fv-clas (v1) and the antigen represented by SEQ ID NO: 30 Crystallization of the complex with the molecule HA peptide (YPYDVPDYA) was performed. However, the 12CA5 antibody-Fv-clasp (v1) and the HA peptide were mixed at a final concentration of 7 mg / mL and 1 mM, respectively. ) (96 conditions), Classics II Suite (QIAGEN) (96 conditions), Wizard Classic 1 & 2 (Rigaku) (96 conditions) were used.
As a result, the complex of 12CA5 antibody-Fv-class (v1) and the antigenic peptide HA peptide in the screening stage was 51 conditions (17.7%) out of a total of 288 conditions using the above-mentioned three types of kits. A crystal was obtained.
When 10 types of crystals of the obtained complex were subjected to an X-ray diffraction experiment in the same manner as in Comparative Example 5, a low diffraction ability of about 4 to 9 mm was exhibited. Therefore, as a result of repeating the examination of crystallization conditions and X-ray diffraction experiments for one of these and optimizing the crystallization conditions, data was successfully collected at 2.2 mm. With respect to this data, the three-dimensional structure was determined by the same method as in Comparative Example 5 (however, PDB ID: 1HIL was used as a search model for molecular replacement for the Fv region). The structure of the complex of the obtained 12CA5 antibody Fv-clasp (v1) and the HA peptide is shown in FIG.
Even if Fv-class (v1) obtained in the E. coli expression system has crystallization ability in the screening stage, the number of screening conditions for obtaining crystals is smaller than that of Fv-class (v2) described later, In addition, it was found that crystals obtained by screening tend to have low diffraction ability (low crystallization ability).

Comparative Examples 7-9 and Example 1 below are based on the structure of Fv-clasp (v1) obtained in Comparative Example 5, and thus improve Fv-clasp (v1). Mutants of Fv-clas (without SS), Fv-clasp (v1 ′), Fv-clasp (v1 ″), and Fv-clasp (v2) described later.
The position of the SS bond in each mutant is shown in Table 1 below.

<Comparative Example 7 Preparation of Fv-clas (no SS) of 2H5 antibody using animal cell expression system>
By the following method, VH-SARAH (containing a 2H5 heavy chain signal sequence at the N terminus and a His tag sequence at the C terminus) for the 2H5 antibody represented by SEQ ID NO: 31. And VL-SARAH for the 2H5 antibody represented by SEQ ID NO: 32 (including a 2H5 light chain signal sequence at the N-terminus and a His tag sequence at the C-terminus. The following may be abbreviated as 2H5 antibody-VL-SARAH. ) Were subjected to (1) gene recombination step, (2) recombinant protein expression step using animal cell expression system, and (3) purification step, and Fv-clas of 2H5 antibody (without SS) ( Hereinafter, a culture supernatant containing 2H5 antibody-Fv-clasp (sometimes abbreviated as SS) may be obtained.
(1) Gene recombination step In place of the nucleic acid sequence encoding VH-SARAH (Y35C) against the P20.1 antibody represented by SEQ ID NO: 23 (including a sequence encoding a mouse nitrogen signal sequence at the N-terminus), A nucleic acid sequence encoding VH-SARAH for the 2H5 antibody shown in SEQ ID NO: 33 (including a signal sequence of 2H5 heavy chain at the N-terminus and a sequence encoding a His tag sequence at the C-terminus) is used, and SEQ ID NO: 25 VL against the 2H5 antibody shown in SEQ ID NO: 34 instead of the nucleic acid sequence encoding VL-SARAH (M24C) against the P20.1 antibody shown in (including a sequence encoding a mouse nitrogen signal sequence at the N-terminus) A nucleic acid sequence encoding SARAH (a signal sequence of 2H5 light chain at the N-terminus and a His tag at the C-terminus) 2H5 antibody-VH-SARAH and 2H5 antibody-VL-SARAH expression recombinant vectors were prepared in the same manner as in Comparative Example 2 except that the sequence encoding the sequence was used.
(2) Recombinant protein expression step using animal cell expression system 2H5 antibody-VH-SARAH expression recombinant vector is used instead of P20.1 antibody-VH-SARAH (Y35C) expression recombinant vector, and The 2H5 antibody-Fv was prepared in the same manner as in Comparative Example 2 except that the recombinant vector for expression of 2H5 antibody-VL-SARAH was used instead of the recombinant vector for expression of P20.1 antibody-VL-SARAH (M24C). The culture supernatant containing -clasp (without SS) was collected.
(3) Purification step From 1 mL of the collected culture supernatant, 20 μL of CNBr-activated Sepharose 4 Fast Flow (GE), to which an eTEV peptide (RENLYFQGKDG) which is an antigen molecule of 2H5 represented by SEQ ID NO: 89 is immobilized, was used. Then, a precipitation reaction was performed by affinity chromatography to elute the protein bound to the carrier to obtain an eluate (hereinafter sometimes abbreviated as an eluate without SS).
In addition, the obtained 2H5 antibody-Fv-clasp (without SS) does not have a new cysteine residue in the Fv region of the 2H5 antibody and the SARAH domain sequence fused thereto, and forms an SS bond between molecules. It is designed not to do so.

<Comparative Example 8 Preparation of 2v5 Antibody Fv-Clasp (v1 ') Using Animal Cell Expression System>
In the same manner as in Comparative Example 7, VH (11C) -SARAH (containing a signal sequence of 2H5 heavy chain at the N-terminus and a His tag sequence at the C-terminus. 2H5 antibody-VH) (11C) -SARAH may be abbreviated), and VL-SARAH (33C) (2H5 light chain signal sequence at the N-terminus and His tag sequence at the C-terminus. 2H5 antibody-VL-SARAH (33C) may be abbreviated as (1) gene recombination step, (2) recombinant protein expression step using animal cell expression system, and (3) purification step And an eluate (hereinafter referred to as v1 ′) containing 2v5 antibody Fv-class (v1 ′) (hereinafter sometimes abbreviated as 2H5 antibody-Fv-class (v1 ′)). Sometimes referred to as the bottom product is) was obtained.
The obtained 2H5 antibody-Fv-clasp (v1 ′) is a cysteine residue of antibody residue number 11 by the Chothia method in the VH region of VH (11C) -SARAH against 2H5 antibody and 2H5 antibody-VL- The SARAH domain of SARAH (33C) is designed to form an SS bond with the 17th cysteine residue from the C-terminus (33th from the N-terminus).

<Comparative Example 9 Preparation of Fv-class (v1 '') of 2H5 Antibody Using Animal Cell Expression System>
In the same manner as in Comparative Example 7, VH (108C) -SARAH (containing a signal sequence of 2H5 heavy chain at the N terminus and a His tag sequence at the C terminus for the 2H5 antibody represented by SEQ ID NO: 37. Hereinafter, 2H5 antibody-VH (108C) -SARAH may be abbreviated), and VL-SARAH (30C) for the 2H5 antibody represented by SEQ ID NO: 38 (including a 2H5 light chain signal sequence at the N-terminus and a His tag sequence at the C-terminus. 2H5 antibody-VL-SARAH (30C) may be abbreviated as (1) gene recombination step, (2) recombinant protein expression step using animal cell expression system, and (3) purification step 2H5 antibody Fv-class (v1 ″) (hereinafter sometimes abbreviated as 2H5 antibody-Fv-class (v1 ″)) v1 ″ eluate is sometimes abbreviated).
The obtained 2H5 antibody-Fv-clasp (v1 ″) is composed of the cysteine residue of antibody residue number 108 by the Chothia method in the VH region of VH (108C) -SARAH for 2H5 antibody and 2H5 antibody-VL. -Designed so that an S-S bond is formed with the 20th cysteine residue from the C-terminus (30th from the N-terminus) in the SARAH domain of SARAH (30C).

<Example 1 Preparation of Fv-class (v2) of 2H5 antibody using animal cell expression system>
In the same manner as in Comparative Example 7, VH (112C) -SARAH (containing a signal sequence of 2H5 heavy chain at the N-terminus and a His tag sequence at the C-terminus. 2H5 antibody-VH) (112C) -SARAH may be abbreviated), and VL-SARAH (37C) (2H5 light chain signal sequence at the N-terminus and His tag sequence at the C-terminus. 2H5 antibody-VL-SARAH (37C) may be abbreviated as (1) gene recombination step, (2) recombinant protein expression step using animal cell expression system, and (3) purification step And an eluate containing the 2H5 antibody Fv-class (v2) (hereinafter sometimes abbreviated as 2H5 antibody-Fv-class (v2)) (hereinafter referred to as v). Sometimes referred to as the eluent there) was obtained.
The obtained 2H5 antibody-Fv-clasp (v2) is a cysteine residue of antibody residue number 112 according to Chothia method in the VH region of VH (112C) -SARAH for 2H5 antibody, and 2H5 antibody-VL-SARAH. The (37C) SARAH domain is designed so that the 13 th cysteine residue from the C terminus (37 th from the N terminus) forms an S—S bond.

<Experimental Example 1 S—S bond forming ability of various Fv-class versions of 2H5 antibody>
Each eluate obtained in Comparative Example 7-9 and Example 1 was subjected to SDS-PAGE under reducing or non-reducing conditions according to a conventional method.
The results are shown in FIG. In the figure, the left side of the gel (lanes 1 to 4) across the central marker lane is an electrophoretic diagram under reducing conditions, and the right side of the gel (lanes 5 to 8) is an electrophoretic diagram under non-reducing conditions. is there. Lane 1 is the SS-free eluate obtained in Comparative Example 7 under reducing conditions, Lane 2 is the v1 ′ eluate obtained in Comparative Example 8 under reducing conditions, and Lane 3 is Comparative Example 9 under reducing conditions. V1 '' eluate obtained in Example 1, Lane 4 is the v2 eluate obtained in Example 1 under reducing conditions, Lane 5 is the eluate without SS obtained in Comparative Example 7 under non-reducing conditions, Lane 6 is the v1 ′ eluate obtained in Comparative Example 8 under non-reducing conditions, Lane 7 is the v1 ″ eluate obtained in Comparative Example 9 under non-reducing conditions, and Lane 8 is conducted under non-reducing conditions. The results of electrophoresis of the v2 eluate obtained in Example 1 are shown respectively.
Under reducing conditions, the SS-free eluate (lane 1), the v1 ′ eluate (lane 2), and the v2 eluate were respectively converted to 2H5 antibody-VH-SARAH or 2H5 antibody-VH-SARAH at around 22 kDa. Mutant bands into which cysteine residues were introduced were obtained, and mutant bands into which cysteine residues were introduced into 2H5 antibody-VL-SARAH or 2H5 antibody-VL-SARAH were obtained in the vicinity of 18 kDa, respectively. On the other hand, the v1 ″ eluate (lane 3) obtained a band corresponding to a mutant in which a cysteine residue was introduced into 2H5 antibody-VH-SARAH or a mutant in which a cysteine residue was introduced into 2H5 antibody-VL-SARAH. I couldn't.
Further, under non-reducing conditions, the SS-free eluate (lane 5) did not give the expected Fv-class dimer band of 2H5 antibody around 35-37 kDa, and v1 ′ eluate (lane 6) In addition to the Fv-clasp (v1 ′) band (around 35 kDa) of the 2H5 antibody that formed an S—S bond, a single 2H5 antibody that did not form an S—S bond near 17 kDa—VH— A band of SARAH and a band of 2H5 antibody-VH-SARAH were obtained, but the v2 eluate (lane 8) had only the band of Fv-class (v2) (37 kDa) of the 2H5 antibody that formed an SS bond. Obtained.
From the results of electrophoresis under reducing conditions, Fv-class (without SS) (Comparative Example 7), Fv-class (v1 ′) (Comparative Example 8), Fv-class (v2) (Comparative Example 10) Was expressed and secreted from the host, respectively, and the binding activity to the antigen molecule was maintained. On the other hand, Fv-clasp (v1 ″) (Comparative Example 9) was found not to be expressed or secreted, or expressed but lost the antigen binding ability.
Furthermore, as a result of electrophoresis under non-reducing conditions, Fv-clasp (v1 ′) produces impurities such as dimers or homodimers that do not form S—S bonds between molecules, whereas Fv-clasp (v2) It was found that only has the ability to form 100% intermolecular S—S bonds.

From the results of the structural analysis of Fv-clasp (v1) shown in Comparative Example 5-6, the present inventors have confirmed that Fv-clasp (v1) and VH-SARAH and VL-SARAH constitute a more stable heterodimer. '), Three different variants of Fv-clasp (v1 ″) and Fv-clasp (v2) were devised.
That is, as a combination of positions of cysteine residues to be introduced into VH-SARAH and VL-SARAH, three new combinations described in Table 1 above {[antibody residue by Chothia method in VH region of VH (11C) -SARAH] A combination of the cysteine residue of group number 11 and the 17th cysteine residue (Fv-clas (v1 ′)) from the C-terminal in the SARAH domain of VL-SARAH (33C) (Fv-clas (v1 ′)), [VH The cysteine residue of antibody residue number 108 by Chothia method in the VH region of (108C) -SARAH and the cysteine residue at the 20th position from the C-terminus (30th position from the N-terminus) in the SARAH domain of VL-SARAH (30C) A combination of groups (Fv-clas (v1 ″)), [VH (112C) -SARAH V The cysteine residue of antibody residue number 112 according to the Chothia method in the region and the 13th cysteine residue (Fv-clasp (v2) from the C-terminus in the SARAH domain of VL-SARAH (37C) )]] Was devised, and the SS bond-forming ability was examined in Experimental Example 1 for them.
However, as shown in Experimental Example 1, surprisingly, among these mutants, Fv-class (v1 ′) does not form a stable SS bond between molecules, and Fv-class (v1 ′) It was found that ') could not maintain good expression level and antigen binding ability.
That is, in order to stabilize the heterodimer by suppressing intramolecular mobility in Fv-clasp and improve the crystallization ability, the position of the S—S bond between heterodimers to be introduced is that of VH (112C) -SARAH. It is necessary to select the amino acid residue of antibody residue No. 112 by Chothia method in the VH region and the 13th amino acid residue from the C-terminus (37th from the N-terminus) in the SARAH domain of VL-SARAH (37C) I understood.

<Example 2 Preparation of Fv-class (v2) of P20.1 antibody using E. coli expression system>
In place of the nucleic acid sequence encoding P20.1 antibody-VH-SARAH (Y35C) shown in SEQ ID NO: 19, the nucleic acid sequence encoding VH (112C) -SARAH against the P20.1 antibody shown in SEQ ID NO: 41 And, instead of the nucleic acid sequence encoding VL-SARAH (M24C) against the P20.1 antibody shown in SEQ ID NO: 20, VL-SARAH (37C) against the P20.1 antibody shown in SEQ ID NO: 42 The VH (112C) -SARAH (hereinafter referred to as P20.A) against the P20.1 antibody represented by SEQ ID NO: 43 is obtained in the same manner as in Comparative Example 1 using an E. coli expression system, except that a nucleic acid sequence that encodes is used. 1 antibody-VH (112C) -SARAH), and the P20.1 antibody shown in SEQ ID NO: 44 VL-SARAH (37C) (hereinafter sometimes abbreviated as P20.1 antibody-VL-SARAH (37C)), (1) gene recombination / expression step and solubilization step, (2) refolding step , (3) Concentration step, and (4) Purification step, Fv-class (v2) against P20.1 antibody (hereinafter sometimes abbreviated as P20.1 antibody-Fv-class (v2)) Obtained.
The results are shown in FIG. In the figure, the left lane is an electrophoretic diagram under reducing conditions, and the right lane is an electrophoretic diagram under non-reducing conditions. Under reducing conditions, P20.1 antibody-Fv-clasp (v2) was obtained by overlapping the bands of VH (112C) -SARAH for P20.1 antibody and VL-SARAH (37C) for P20.1 antibody around 18 kDa. Under non-reducing conditions, a single band of P20.1 antibody-Fv-class (v2) was obtained around 35 kDa. Fv-clasp (v2) is a highly purified recombinant protein that forms a heterodimer correctly between VH (112C) -SARAH and VL-SARAH (37C) in the E. coli expression system and forms an SS bond. It turns out that it is obtained.
In addition, since the obtained P20.1 antibody-Fv-clasp (v2) is purified by a resin in which the C8 peptide as an antigen molecule is immobilized in the above purification process, it has activity as an anti-C8 peptide antibody. Obviously.
That is, it was found that Fv-clasp (v2) can be easily produced regardless of the type of host cell to be expressed and has antigen binding activity (Examples 1 and 2).

<Example 3-10 Preparation of Fv-clasp (v2) using E. coli expression system>
In the same manner as in Example 2, instead of VH (112C) -SARAH against P20.1 antibody, 93201 antibody, TS2 / 16 antibody, SG / 19 antibody, t8E4 antibody, 9E10 antibody, 12CA5 antibody, t1E4 antibody, and Instead of VH (112C) -SARAH for NZ-1 antibody and VL-SARAH (37C) for P20.1 antibody, 93201 antibody, TS2 / 16 antibody, SG / 19 antibody, t8E4 antibody, 9E10 antibody, 12CA5 antibody VL-SARAH (37C) against t1E4 antibody and NZ-1 antibody, respectively, (1) gene recombination / expression step and solubilization step, (2) refolding step, (3) concentration step, and (4) Fv-class (v2) for each antibody (hereinafter, the name of each antibody) Sometimes abbreviated -fv-clasp and (v2) is) was prepared.
The sequences of the nucleic acid used and the prepared protein are shown in Table 2 and Table 3, respectively.

結果を図８に示す。図中、左側のレーンが還元条件下での電気泳動図、右側のレーンが非還元条件下での電気泳動図である。８種類の抗体のＦｖ−ｃｌａｓｐ（ｖ２）のすべてにおいて、実施例２で得られたＰ２０．１抗体−Ｆｖ−ｃｌａｓｐ（ｖ２）と同様に、還元条件下では各抗体に対するＶＨ（１１２Ｃ）−ＳＡＲＡＨ及びＶＬ−ＳＡＲＡＨ（３７Ｃ）に相当するバンドが１８−２０ｋＤａ付近に得られ、また、非還元条件下では３５−３７ｋＤａ付近に各抗体に対するＦｖ−ｃｌａｓｐ（ｖ２）のバンドが一本得られた。
すなわち、大腸菌発現系で調製したこれら８種の抗体のＦｖ−ｃｌａｓｐ（ｖ２）は、いずれもＶＨ（１１２Ｃ）−ＳＡＲＡＨとＶＬ−ＳＡＲＡＨ（３７Ｃ）の間で正しくヘテロダイマーを形成してＳ−Ｓ結合を形成し、高い純度で精製できることが判った。

The results are shown in FIG. In the figure, the left lane is an electrophoretic diagram under reducing conditions, and the right lane is an electrophoretic diagram under non-reducing conditions. In all of the 8 types of Fv-class (v2) of the antibody, the VH (112C) -SARAH for each antibody under reducing conditions, similar to the P20.1 antibody-Fv-class (v2) obtained in Example 2 And a band corresponding to VL-SARAH (37C) was obtained in the vicinity of 18-20 kDa, and one Fv-class (v2) band for each antibody was obtained in the vicinity of 35-37 kDa under non-reducing conditions.
That is, the Fv-clasp (v2) of these 8 types of antibodies prepared in the E. coli expression system correctly forms heterodimers between VH (112C) -SARAH and VL-SARAH (37C), and thus S-S It was found that a bond was formed and could be purified with high purity.

<Experimental Example 2 Confirmation of Antigen Binding Activity of Fv-clasp (v2)>
It was confirmed by gel filtration chromatography peak shift or biolayer interferometry that Fv-clasp (v2) against various antibodies obtained in Example 3-10 had antigen binding activity.
For Fv-class (v2) against 93201 antibody, TS2 / 16 antibody, SG / 19 antibody and t8E4 antibody, the respective antigen molecule [SORLA Vps10p domain represented by SEQ ID NO: 77 for Fv-class (v2) against 93201 antibody , Integrin α6β1, t8E4 antibody comprising α6 chain represented by SEQ ID NO: 78 and β1 chain represented by SEQ ID NO: 79 for Fv-class (v2) against TS2 / 16 antibody and Fv-class (v2) against SG / 19 antibody It was confirmed by the peak shift of gel filtration chromatography that a stable complex was formed with Fv-clasp (v2) against HGF shown in SEQ ID NO: 80].
The results are shown in FIG. Since Fv-class (v2) for any antibody shifted its peak due to binding to the antigen protein, it was found that Fv-class (v2) for various antibodies had antigen-binding activity.
The Fv-class (v2) for the 9E10 antibody and the Fv-class (v2) for the 12CA5 antibody are represented by the T4L protein to which the Myc tag represented by SEQ ID NO: 81 as the respective antigen molecule is added, or SEQ ID NO: 82. The binding to the HA-tagged T4L protein was analyzed in real time by biolayer interferometry using Octet Red96 (Primetech) and compared with the binding ability of the original antibodies (9E10 antibody and 12CA5 antibody) to the antigen molecule. .
The results are shown in FIG. It was found that Fv-clasp (v2) for any antibody has the same level of antigen binding activity (antigen affinity) as the original antibody.
As for Fv-clasp (v2) against NZ-1 antibody and Fv-clasp (v2) against t1E4 antibody, in Example 14 and Example 17 described later, crystal structure analysis in the state of complex with each antigen It was directly confirmed that a 1: 1 complex with the antigen was formed in the crystal. From these results, it was found that Fv-clasp (v2) against various antibodies has antigen binding activity.

<Example 11 Crystal structure analysis of Fv-clasp (v2) against TS2 / 16 antibody>
Using Fv-clasp (v2) (hereinafter sometimes abbreviated as TS2 / 16 antibody-Fv-clas (v2)) against the TS2 / 16 antibody prepared in Example 4, the same method as in Comparative Example 4 was used. Crystallization of TS2 / 16 antibody-Fv-clasp (v2) was performed.
However, 4.9 mg / mL TS2 / 16 antibody-Fv-clasp (v2) solution was used as a crystallization sample, and the screening kits were Classics Neo Suite (manufactured by QIAGEN) (96 conditions), Wizard Classic 1 & 2 (Rigaku) (96 conditions).
As a result, in the screening stage, crystals were obtained under one of 192 conditions (0.5%). About the obtained crystal | crystallization, the X-ray-diffraction data was measured by the method similar to the comparative example 5, data were obtained with 2.04 resolution | decomposability, and the structure was able to be determined.
The results are shown in FIG. Fv-clasp (v2) was found to have crystallization ability. In general, full-length antibodies usually do not yield crystals, and conventional fragment antibodies (Fab fragments or single-chain antibodies (scFv) fragments) often do not yield crystals, so the three-dimensional structure analysis of the antigen-determining site is unknown. Antibodies are present. Since Fv-clasp (v2) has crystallization ability and is easy to crystallize, it was proved to be particularly useful in the three-dimensional structure analysis of the antigen determining site of the antibody.

<Example 12-14 Crystal structure analysis of complex of various Fv-clasp (v2) and antigen>
Complex (Example 12) of P20.1 antibody-Fv-clasp (v2) prepared in Example 2 and C8 peptide which is an antigen molecule of P20.1 antibody represented by SEQ ID NO: 21, prepared in Example 8 12CA5 antibody-Fv-clasp (v2) and an HA peptide that is an antigen molecule of the 12CA5 antibody represented by SEQ ID NO: 30 (Example 13), NZ-1 antibody-Fv-clas prepared in Example 10 The complex (Example 14) of (v2) and the PA peptide that is the antigen molecule of the NZ-1 antibody represented by SEQ ID NO: 83 was crystallized by the same method as in Comparative Example 4.
However, for the complex of P20.1 antibody-Fv-clasp (v2) and C8 peptide, a mixture of the two so that the final concentration of each was 11 mg / mL and 2 mM was used as a crystallization sample, and the screening kit was , Classics Neo Suite (manufactured by QIAGEN) (96 conditions), Classics II Suite (manufactured by QIAGEN) (96 conditions), Wizard Classic 1 & 2 (manufactured by Rigaku) (96 conditions), JCSG + (Molecular Dimes) 96 conditions). Regarding the complex of 12CA5 antibody-Fv-clasp (v2) and HA peptide, a mixture of both was used as a crystallization sample so that the final concentrations thereof were 7 mg / mL and 1 mM, and the screening kit was Classics Neo Suite. (QIAGEN) (96 conditions), Classics II Suite (QIAGEN) (96 conditions), Wizard Classic 1 & 2 (Rigaku) (96 conditions) were used. As for the complex of NZ-1 antibody-Fv-clasp (v2) and PA peptide, a mixture of the two so as to have a final concentration of 3.8 mg / mL and 1 mM is used as a crystallization sample. , Classics Neo Suite (manufactured by QIAGEN) (96 conditions), Wizard Classic 1 & 2 (manufactured by Rigaku) (96 conditions) were used.
As a result, at the screening stage, for the complex of P20.1 antibody-Fv-class (v2) and C8 peptide, 35 out of 384 conditions (9.1%), 12CA5 antibody-Fv-class (v2) For the complex with HA peptide, 74 conditions (25.7%) out of 288 conditions, and for the complex of NZ-1 antibody-Fv-class (v2) and PA peptide, 6 conditions in 192 conditions (3. 1%) crystals were obtained respectively.
About the complex of 12CA5 antibody-Fv-clasp (v2) and HA peptide and the complex of NZ-1 antibody-Fv-clasp (v2) and PA peptide, using the crystals obtained by screening, Comparative Example 5 X-ray diffraction data was measured by the same method as described above, and data was obtained with a resolution of 2.45 mm and a resolution of 2.15 mm, respectively. In addition, for the complex of P20.1 antibody-Fv-clasp (v2) and C8 peptide, the crystallization conditions were optimized based on the crystal conditions obtained by screening, and the resulting crystals were The X-ray diffraction data was measured using this, and data with a resolution of 1.17 mm was obtained. The structure was determined by the same method as in Comparative Example 5 using the obtained X-ray diffraction data. However, for the P20.1 antibody-Fv-class (v2), the search model for molecular substitution for the Fv region is the structure of the complex of P20.1 antibody-Fv-class (v1) and C8 peptide, 12CA5 antibody-Fv -Clasp (v2): 12CA5 antibody -Fv-clasp (v1) and HA peptide complex structure, NZ-1 antibody -Fv-clasp (v2) NZ-1 Fab structure (PDB ID: 4YO0) It was used.
The results are shown in FIGS.

Fv-clasp (v1) and Fv-clasp (v2) will be considered in comparison with the following.
As a result of Comparative Example 4, the complex of P20.1 antibody-Fv-class (v1) and C8 peptide obtained in the E. coli expression system was not able to obtain crystals at all, although 480 conditions were examined. Crystals were obtained with a probability of 3.6% for the first time after being prepared in the cell expression system, whereas as a result of Example 12, the complex of P20.1 antibody-Fv-class (v2) and C8 peptide was Crystals obtained with the E. coli expression system were obtained under 35 conditions (9.1%) out of 384 conditions, and crystals with optimized crystallization conditions gave very high resolution data with 1.17Å resolution. It was.
In addition, as a result of Comparative Example 6, crystals of 12CA5 antibody-Fv-clasp (v1) and the HA peptide in the screening stage were obtained under 51 conditions (17.7%) out of 288 conditions. Although only low resolution data of about 4 to 9 resolution was obtained, the same conditions (protein concentration, antigen concentration, screening used) as the complex of 12CA5 antibody-Fv-class (v1) and HA peptide Although the screening was carried out with the type of kit), as a result of Example 13, the complex of 12CA5 antibody-Fv-class (v2) and the HA peptide in the screening stage was 74 conditions (25.7) in 288 conditions. %) Crystals were obtained, and high resolution data with a resolution of 2.45 mm was obtained from the crystals.
From these results, although Fv-class (v1) obtained in an E. coli expression system may not have crystallization ability, Fv-class (v2) crystallizes regardless of the host at the time of expression. It was found that a product having the ability was obtained. Since E. coli is easier to handle than animal cells, Fv-class (v2) is more useful than Fv-class (v1) particularly in that an E. coli expression system has crystallization ability.
Moreover, even when Fv-class (v1) obtained in an E. coli expression system has crystallization ability, Fv-class (v2) has a higher crystallization ability than Fv-class (v1). It was found that high resolution crystals can be obtained (easily crystallized).

<Example 15-17 Crystal structure analysis of complex of various Fv-clasp (v2) and hardly crystalline antigen>
Regarding the Vps10p domain of SORLA (neural sorting receptor), integrin α6β1, and HGF (hepatocyte growth factor), which is a hardly crystalline protein, the Vps10p represented by SEQ ID NO: 77 and the 93201 antibody-Fv-clasp prepared in Example 3 ( v2) (Example 15), integrin α6β1 represented by SEQ ID NO: 78 and SEQ ID NO: 79 and TS2 / 16 antibody-Fv-clas (v2) prepared in Example 4 (Example 16) The complex (Example 17) of HGF represented by SEQ ID NO: 80 and the t1E4 antibody-Fv-class (v2) prepared in Example 9 was crystallized by the same method as in Comparative Example 4.
However, for the complex of Vps10p and 93201 antibody-Fv-clasp (v2), those with concentrations of 3 mg / mL and 1.5 mg / mL were used as crystallization samples, and the screening kit was a Classics Neo Suite for the former. (QIAGEN) (96 conditions), and for the latter, Classics Neo Suite (QIAGEN) (96 conditions) and Wizard Classic 1 & 2 (Rigaku) (96 conditions) were used. For the complex of integrin α6β1 and TS2 / 16 antibody-Fv-clasp (v2), a crystallization sample was used at a concentration of 9.2 mg / mL, and the screening kit was Classics Neo Suite (manufactured by QIAGEN) ( 96 conditions), Wizard Classic 1 & 2 (manufactured by Rigaku) (96 conditions), and ProPlex (manufactured by Molecular Dimensions) (96 conditions). For the complex of HGF and t1E4 antibody-Fv-class (v2), a crystallization sample was used at a concentration of 9.7 mg / mL, and the screening kit was Classics Neo Suite (manufactured by QIAGEN) (96 conditions) , Classics II Suite (manufactured by QIAGEN) (96 conditions), Wizard Classic 1 & 2 (manufactured by Rigaku) (96 conditions), ProPlex (manufactured by Molecular Dimensions) (96 conditions), JCSG + (manufactured by Molecular Dimension 96) Condition) was used.
As a result, at the screening stage, for the complex of Vps10p and 93201 antibody-Fv-clasp (v2), 11 out of 288 conditions (3.8%), integrin α6β1 and TS2 / 16 antibody-Fv-clasp (v2) ) For 2 complexes (0.7%) in 288 conditions, and for the complex of HGF and t1E4 antibody-Fv-class (v2), 1 condition (0.2%) in 480 conditions Crystals were obtained for each.
About the obtained crystal | crystallization of each composite_body | complex, the X-ray-diffraction data were measured by the method similar to the comparative example 5, and the X-ray crystal structure analysis of the said composite_body | complex was performed.
The results are shown in FIGS. In the screening stage, the complex of Vps10p and 93201 antibody-Fv-class (v2) has a resolution of 3.2, and the complex of integrin α6β1 and TS2 / 16 antibody-Fv-class (v2) has a resolution of 3.37. As for the complex of HGF and t1E4 antibody-Fv-clas (v2), data was obtained at a resolution of 4.1Å, and the crystal structure could be determined.
Furthermore, when the crystallization conditions were optimized, the complex of Vps10p and 93201 antibody-Fv-class (v2) was found to have a resolution of 2.59Å, integrin α6β1 and TS2 / 16 antibody-Fv-class (v2) With respect to the complex, data was finally obtained at a resolution of 3.05 mm, and the crystal structure could be determined.

Below, the function which accelerates | stimulates crystallization of Fv-clasp (v2) is considered.
Integrins are multidomain glycoproteins, and the extracellular domain is known to undergo very large structural changes in conjunction with signal transduction, and it has been thought that crystallization alone is difficult. Thus, the present inventors have attempted crystallization using a plurality of antibody Fabs containing TS2 / 16, but no crystals have been obtained. However, as described above, by using the complex of TS2 / 16 antibody with Fv-class (v2), data was obtained at a resolution of 3.37 mm even in the screening stage, and the crystal structure could be determined.
In addition, the inventors of the present application tried to analyze the crystal structure of Fab of HGF and t1E4 antibody in the past, but only a crystal having a diffraction ability of about 9 mm at the maximum was obtained. However, as described above, by using Fv-clasp (v2), the crystal structure could be determined with a resolution of 4.1Å even at the screening stage.
From these results, it has been clarified that Fv-clasp (v2) has a function of promoting crystallization when it forms a complex with a protein, particularly a hardly crystalline protein. As useful.
In addition, the inventors of the present invention have previously performed a crystal structure of a complex of Vps10p and a propeptide (Kitago, Y., et al. Structural basis for amyloidogenic peptide recognition by LS. Nature Str. 199-206. (2015)), however, it took about 1 to 2 months for the crystals of the complex to precipitate, and the frequency of obtaining crystals with a quality that could withstand the analysis was very low. However, as described above, by using Fv-clasp (v2), the crystal structure can be determined at a resolution of 3.2 mm and at a resolution of 2.59 mm after optimization of the crystallization conditions, even in the screening stage. These crystals could be obtained with good reproducibility in just one day.
That is, Fv-clasp (v2) has been shown to have a function of promoting crystallization of proteins, particularly difficult-to-crystal proteins, with good reproducibility and in a short time, and is useful as a fragment antibody for promoting crystallization. It turns out that there is.

<Experimental example 3 Fv-clasp (v2) thermal stability>
The thermal stability of scFv and Fv-class (v1) and Fv-class (v2), which are conventional single chain antibodies, was compared by the following method.
The scFv (single-chain Fv) of the P20.1 antibody, which is a conventional fragment antibody, was prepared by E. coli expression and unwinding by the method described in Japanese Patent No. 5257997. The scFv of the NZ-1 antibody was prepared by the same method as the single chain antibody (scFv) of the P20.1 antibody except that the sequence shown in SEQ ID NO: 84 was used.
In addition, Fv-class (v1) for NZ-1 antibody and Fv-class (v1) for SG / 19 antibody were determined by the same method as in Comparative Example 3 and VH- against NZ-1 antibody represented by SEQ ID NO: 85. SARAH (Y35C) and VL-SARAH (M24C) against the NZ-1 antibody, as shown in SEQ ID NO: 86, and VH-SARAH (Y35C) and as shown in SEQ ID NO: 88, against the SG / 19 antibody, as shown in SEQ ID NO: 87 VL-SARAH (M24C) against SG / 19 antibody is subjected to (1) gene recombination / expression step and solubilization step, (2) refolding step, (3) concentration step and (4) purification step, respectively. Prepared.
The prepared scFv of P20.1 antibody, scFv of NZ-1 antibody, Fv-class (v1) against NZ-1 antibody, Fv-class (v1) against SG / 19 antibody, P20.1 obtained in Comparative Example 1 Antibody-Fv-class (v1), 12CA5 antibody-Fv-class (v1) obtained in Comparative Example 3, P20.1 antibody-Fv-class (v2) prepared in Example 2, prepared in Example 5 Thermally using SG / 19 antibody-Fv-clasp (v2), 12CA5 antibody-Fv-clasp (v2) prepared in Example 8, and NZ-1 antibody-Fv-clasp (v2) prepared in Example 10 A shift assay was performed to determine each Tm value.
In the thermal shift assay, SYPRO Orange (manufactured by Thermo Fisher Scientific) is added to each sample, and the change in fluorescence value is measured in real time when the temperature is raised from 25 ° C to 85 ° C. The state was monitored, and the Tm value [° C.] was calculated from the obtained denaturation curve. The results are shown in Table 4 below. All units in the table are [° C.].

これらの実験結果から、Ｆｖ−ｃｌａｓｐ（ｖ２）は、従来のフラグメント抗体であるｓｃＦｖやＦｖ−ｃｌａｓｐ（ｖ１）に比べＴｍ値が高く、その差は６−１１℃にも達することが判った。すなわち、Ｆｖ−ｃｌａｓｐ（ｖ２）は、従来のフラグメント抗体よりもはるかに熱安定が高い為、冷蔵あるいは室温での保存安定性も高い。この性質により、Ｆｖ−ｃｌａｓｐ（ｖ２）は、フラグメント抗体、及びタンパク質結晶化促進用フラグメント抗体としてはもちろん、それ以外の用途でも極めて有利な性質を有することが期待される。

From these experimental results, it was found that Fv-clasp (v2) had a higher Tm value than scFv and Fv-clasp (v1), which are conventional fragment antibodies, and the difference reached 6-11 ° C. That is, since Fv-clasp (v2) is much more thermally stable than conventional fragment antibodies, it is also highly refrigerated or stored at room temperature. Due to this property, Fv-clasp (v2) is expected to have extremely advantageous properties not only as a fragment antibody and a fragment antibody for promoting protein crystallization but also in other uses.

Claims (7)

(A) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the heavy chain domain (VH region) of the antibody, and the VH region has been mutated to the amino acid residue of antibody residue number 112 by the Chothia method to cysteine A peptide (VH (112C) -SARAH);
(B) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody, wherein the SARAH domain has a 13th amino acid residue mutated to cysteine from the C-terminus (VL -A fragment antibody consisting of a complex with SARAH (37C)),
(C) A fragment antibody in which VH (112C) -SARAH and VL-SARAH (37C) are bound by a disulfide bond between the two cysteines. The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1 to 8, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9 to 16. 2. The fragment antibody according to 1. The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1-2, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9-10. 2. The fragment antibody according to 1. (A) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the heavy chain domain (VH region) of the antibody, and the VH region has been mutated to the amino acid residue of antibody residue number 112 by the Chothia method to cysteine A peptide (VH (112C) -SARAH);
(B) A peptide in which the N-terminus of the SARAH domain is bound to the C-terminus of the light chain domain (VL region) of the antibody, wherein the SARAH domain has a 13th amino acid residue mutated to cysteine from the C-terminus (VL -A fragment antibody for promoting protein crystallization comprising a complex with SARAH (37C)),
(C) A fragment antibody for promoting protein crystallization, in which VH (112C) -SARAH and VL-SARAH (37C) are bound by a disulfide bond between the two cysteines. The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1 to 8, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9 to 16. 4. The fragment antibody for promoting protein crystallization according to 4. The SARAH domain in the VH (112C) -SARAH is represented by SEQ ID NOs: 1-2, and the SARAH domain in the VL-SARAH (37C) is represented by SEQ ID NOs: 9-10. 4. The fragment antibody for promoting protein crystallization according to 4. A method for crystallizing a protein using the fragment antibody shown in any one of claims 1 to 3.

Images